Data communications and enterprise networking

Transcription

Data communications and
enterprise networking
Volume 1
P. Tarr
CO2222
2005
Undergraduate study in
Computing and related programmes
This is an extract from a subject guide for an undergraduate course offered as part of the
University of London International Programmes in Computing. Materials for these programmes
are developed by academics at Goldsmiths.
For more information, see: www.londoninternational.ac.uk
This guide was prepared for the University of London International Programmes by:
P. Tarr
This is one of a series of subject guides published by the University. We regret that due to pressure of work the author is
unable to enter into any correspondence relating to, or arising from, the guide. If you have any comments on this subject
guide, favourable or unfavourable, please use the form at the back of this guide.
University of London International Programmes
Publications Office
32 Russell Square
London WC1B 5DN
United Kingdom
www.londoninternational.ac.uk
Published by: University of London
© University of London 2005
The University of London asserts copyright over all material in this subject guide except where otherwise indicated. All rights
reserved. No part of this work may be reproduced in any form, or by any means, without permission in writing from the
publisher. We make every effort to respect copyright. If you think we have inadvertently used your copyright material, please
let us know.
Contents
Contents
Introduction
1.1 Aims and objectives
1.2 Learning outcomes
1.3 Course outline
1.4 How to use this subject guide
1.5 Reading list
1.6 Useful web links
1.7 Study time
1.8 Examination
Chapter 2: Basic concepts
Introduction
Further reading
2.1 Data communications
2.2 Shannon’s Communication Model
2.3 Channels
2.4 Networks
Specimen examination question
Learning outcomes
Chapter 3: Network architecture
Introduction
Further reading
3.1 Layered architectures
3.2 Reference models
3.3 Network standards
3.4 Network devices
3.5 Network protocols
Learning outcomes
Chapter 4: The application layer
Introduction
Further reading
4.1 Services
4.2 Interfaces
4.3 Functions
4.4 Virtual terminal protocols (Telnet, SSH and VT)
4.5 Web protocols (HTTP and HTML)
4.6 Mail protocols (SMTP, MIME, POP, IMAP and MOTIS)
4.7 File transfer and access protocols (FTP, TFTP, NFS and FTAM)
4.8 Directory protocols (DNS, X.500 and LDAP)
4.9 Network Management Protocols (SNMP and CMIP)
Learning outcomes
Chapter 5: The transport layer
Introduction
Further reading
5.1 Services
5.2 Interfaces
5.3 Functions
5.4 User Datagram Protocol (UDP)4
1
1
1
2
3
3
3
4
4
5
5
5
5
7
8
12
16
16
17
17
17
17
19
19
20
23
27
28
29
29
29
29
30
30
34
35
37
40
42
43
44
45
47
47
47
47
48
49
53
i
Data communications and enterprise networks
5.5 Transmission Control Protocol (TCP)
5.6 Sequenced Packet Exchange (SPX)
5.7 Transport Layer Security (TLS)
Learning outcomes
Chapter 6: The network layer
Introduction
Further reading
6.1 Services
6.2 Interfaces
6.3 Functions
6.4 Internet Protocol version 4 (IPv4)
6.5 Internet Control Message Protocol (ICMP)
6.6 Internet Group Management Protocol (IGMP)
6.7 IP Security Protocols (IPsec)
6.8 Address Resolution Protocols (ARP19 and RARP20)
6.9 Internet Protocol Version 6 (IPv6)
6.10 ICMP Version 6 (ICMPv6)
Learning outcomes
Chapter 7: The data link layer
Further reading
7.1 Services
7.2 Interfaces
7.3 Functions
7.4 Binary Synchronous Communications (BSC or BiSync)
7.5 High-level Data Link Control (HDLC)
7.6 Point-to-Point Protocol (PPP)
7.7 Logical Link Control (LLC)
7.8 Ethernet Media Access Control (IEEE 802.3)17
Learning outcomes
Chapter 8: The physical layer
Introduction
Further reading
8.1 Services
8.2 Interfaces
8.3 Functions
8.4 Dial-up modems
8.5 Digital Subscriber Line (DSL)
8.6 Cable modems
Learning outcomes
Appendix A
Specimen examination paper
Appendix B
Model answers and hints
A: Specimen examination paper
Appendix C
List of acronyms
Comment form
ii
53
58
58
60
60
61
61
61
61
63
63
65
71
73
73
74
74
77
77
77
79
79
79
79
79
92
92
93
94
95
96
96
97
97
97
97
97
98
113
114
115
116
116
117
117
119
119
122
125
125
130
Chapter 1: Introduction
Introduction
1.1 Aims and objectives
The CIS222 module aims to provide you with a good grounding in data
communications and enterprise networking. It is a level two course, building
on material taught in level one modules. The course concentrates on breadth
of understanding rather than depth, and attempts to cover a wide range of
networking topics by means of a structured high-level approach. If you desire
a deeper understanding of any of the topics covered in this subject guide, you
should refer to the recommended texts.
Networks have become increasingly important in recent years. Hardly any
applications are now stand-alone. Most applications communicate over a
network, either between a client PC and a server in the same building over a
Local Area Network or between computers in different cities, countries or
continents over a Wide Area Network. The study of networks and
applications that make use of networks is now an important part of any
undergraduate computer science programme. An understanding of how
networks operate will be useful in any computing career you may choose to
follow.
The main objective is for you to build a logical framework in which
networking topics can be studied, analysed and understood. This will stand
you in good stead if you subsequently work in the computer or networking
industry and have to design, develop or manage systems that make use of
networks. This subject, like many in Computer Science, develops rapidly and
new network protocols and technologies are emerging all the time. If you
have built a framework which you then can use to analyse and understand
these new developments, this will ultimately be of more use than a detailed
understanding of current technologies.
A further objective is to help you be able to make informed choices amongst
the many competing technologies that are available in the marketplace, if in
a subsequent career you are required to make such choices.
The CIS222 module replaces the CIS208 module entitled
Telecommunications and Computer Communications. Some of the
material in the CIS208 module has become very dated, as new technologies
have emerged and data transmission speeds have increased way beyond
what was imagined possible ten years ago. Certain promising technologies
which were considered important in the CIS208 module have also failed to
come into widespread use, mainly as a result of the phenomenal success of
the Internet which did not have a very high profile in CIS208. As a result of
this, the new CIS222 module unlike CIS208 is very much focused on the
Internet and its protocols.
1.2 Learning outcomes
On completion of this course and the relevant reading, you should be able to:
• explain the operation of different data communications protocols and
their roles within layered network architectures
• explain the need for standards and outline the roles of the various
standards bodies responsible for data communications standardisation
1
Data communications and enterprise networking – volume 1
• describe the advantages and disadvantages of different network
topologies and technologies
• demonstrate skills in using simple network management tools such as
ping, traceroute and netstat and in using spreadsheets to analyse and
display results.
1.3 Course outline
The CIS222 module consists of two distinct parts which correspond to the
two volumes of the subject guide. The first part of the course is called ‘Data
Communications’ and is an introduction to the terms, concepts and network
architectures required to understand how data is transmitted through
communications channels and networks. This, of necessity, requires a
technical approach and some knowledge of the physics of transmission as
well as the study of network architectures and protocols. The second part of
the module is called ‘Enterprise Networking’ and is more concerned with the
design and management of networks used by businesses and other large
organisations. It therefore concentrates more on the business and
management issues that arise.
This volume of the subject guide can be regarded as a comparative
technical study of layered network protocols1. This volume contains eight
chapters. This chapter will introduce the subject. The second chapter will
introduce the terms and concepts that are used to describe data
communications systems and the third chapter will describe network
architectures and reference models. The next five chapters will examine
protocols in the five layers of the hybrid reference model. There will be
chapters on the application, transport, network, data link and physical
layers. This approach gives equal status to each layer and ensures a
consistent structure to each chapter. Each chapter has sections on
services, interfaces, functions and protocols.
In this part of the subject, we will follow an unconventional approach.
Material will be presented in a top-down fashion. Most courses in the
subject start with the physics of data transmission and build the other
communications layers progressively on top of this foundation.
Eventually, after all the other layers have been studied, the application
layer is then examined. While this is a natural approach for electrical
engineering students who will have a good prior knowledge of physics,
it is less satisfactory for many computer science students, some of whom
have very little knowledge of physics but virtually all of whom will have
a good knowledge of network applications such as email and the world
wide web. The top-down approach therefore starts with applications
with which you are familiar, and then examines progressively the other
communication layers, ending with the physical layer which many
students find the most difficult. The main advantage of this top-down
approach is that you will see the relevance of the material in the context
of your existing knowledge and will not be put off the subject at the
initial stage because you find the physics difficult and cannot see its
relevance.
This subject guide will contain practical activities that can be carried out
using simple network diagnostic tools like ping and traceroute, which are
supplied with operating systems such as Windows, MacOS and Linux/Unix.
These activities can be carried out in a laboratory or at home. Knowledge of
how to use these simple tools will enable you to investigate network
problems on your own.
2
1
The rules which are used to
encode and transmit data across
networks and the technology that
employ these protocols.
Chapter 1: Introduction
1.4 How to use this subject guide
This subject guide is intended to provide a logical structure in which to study
the subject of Enterprise Networking. It is not intended to replace your need
to read around the subject to improve your understanding. You may wish to
follow some of the further reading referenced at the start of each chapter. An
alternative or supplement to the further reading is to follow the links on the
course web site2 to study the topics in each chapter.
How to best use this subject guide will depend on your personal approach to
study and whether you are studying on your own or at an institution.
2
http://doc.gold.ac.uk/mas01pt/cis222/usefullinks.htm.
One suitable approach would be to start with reading a chapter of the subject
guide, followed by some of the further reading, if any of the material has not
been understood or more information is desired. Then attempt the sample
examination questions at the end of the chapter, which can be checked with
the model answers and hints in Appendix B, before reading the learning
outcomes. If you feel that these have not been fully achieved, go back to the
further reading or to the web links, in order to make sure that you have fully
understood each topic. You should also carry out the activities in each
chapter, as they provide further insight to the topics being studied.
This subject guide makes use of a large number of acronyms. When a new
acronym is introduced, it is expanded in full. Subsequent references will
often just use the acronym. A list of acronyms used in the subject guide can
be found in Appendix C.
1.5 Reading list
Because of the very varied material presented in this course, there is no one
book that covers the whole course (or even the majority of it) and that can be
recommended for essential reading. You may wish to obtain your own copy
of one of the general networking books listed below or use library books and
websites for further reading material.
Tanenbaum, Andrew S. Computer Networks. (Prentice Hall International),
fourth edition, 2002.
Forouzan, Behrouz, A. Data Communications and Networking. (McGraw Hill),
third edition, 2003.
Kurose, James F. and Ross, Keith W. Computer Networking. (Addison-Wesley),
third edition, 2005.
Stallings, William Data and Computer Communications. (Prentice Hall
International), seventh edition, 2003.
1.6 Useful web links
The world wide web is a very dynamic medium and publishing a large
number of web links in a subject guide is likely to result in the frustration of a
‘File Not Found’ response at some point in the future. Instead, an up-to-date
list of useful links relevant to the course will be maintained on the CIS222
course web site at
http://doc.gold.ac.uk/~mas01pt/cis222/study/volume_2.htm.
In the event of this URL changing during the life of the subject guide, please
follow the links to the author’s home page from the staff page of the
Goldsmiths’ Department of Computing web site, and then follow the link to
this page.
3
Two links are particularly useful:
http://www.rfc-editor.org/rfc.html – contains the full text of all the Internet
standards which are known as Requests for Comments (RFCs).
http://www.protocols.com/protocols.htm – contains descriptions of most
network protocols.
1.7 Study time
If you are studying this module full time then it should take approximately
one quarter of your total study time for an academic year. In a typical
institution you will probably spend about four hours per week on each
module during term time in formal teaching. It is recommended that you
spend at least three hours per week in reading and private study. You should
also on average spend a further three hours per week in attempting sample
examination questions at the end of each chapter and in doing the activities,
plus formal coursework, in the lab or at home. If you are studying entirely on
your own, then you should aim to spend an additional four hours per week in
private study, to compensate for the hours that students in institutions
receive formal teaching.
You should aim to spend a total of 300 hours on the whole module, including
formal teaching. This total should also include time spent revising during
reading weeks, vacations and prior to examinations. This times is applicable
to an average student who aims to do well in the course. Some students who
work more slowly may need to devote more time than this.
In order to complete this volume of the study guide in one (ten-week) term,
you should aim to complete one chapter per week. This will leave two weeks
spare for consolidation at the end or to allow for some slippage.
If you are studying part-time, over a longer period than one term, then you
will have to adjust this recommended study time accordingly. Revision for
examinations should be in addition to the above.
1.8 Examination
Important: the information and advice given in the following section are
based on the examination structure used at the time this guide was written.
However, the University can alter the format, style or requirements of an
examination paper without notice. Because of this we strongly advise you to
check the rubric/instructions on the paper you actually sit.
The examination will be a single three-hour written paper, usually sat in
May, which will consist of a total of six questions. Three questions will be on
the first half of the course (Data communications) and three on the second
half of the course (Enterprise networking). You will be expected to answer
four questions in all: two from the first half of the paper and two from the
second half of the paper. Specimen examination questions can be found at
the end of each chapter and a complete specimen examination paper can be
found in Appendix A. Some model answers and hints can be found in
Appendix B.
The overall mark for the course will be calculated from the examination
results and the coursework marks.
4
Introduction
In this chapter we will take a brief look at the history of and define some
basic terms used in data communications. We will examine Shannon’s model
of communications and classify different types of communications channels
and study their characteristics. Finally, we shall study different types of
networks and topologies.
Further reading
third edition, (2003). Chapters 1.1–1.3, 3.5, 3.6.
Kurose, James F. and Keith W. Ross Computer Networking. (Addison-Wesley),
third edition (2005). Chapters 1.2–1.3, 1.6.
International), seventh edition (2003). Chapter 1.1, 1.3, 3.4, 15.2, App 3A.
fourth edition, (2002). Chapters 1.1, 1.2, 2.1.3, 2.5.3.
2.1 Data communications
2.1.1 Brief history
Data communications is becoming an increasingly important part of
everyday life, with the explosive growth of the Internet, third generation
mobile networks that support multi-media applications and the increasing
need for enterprises to communicate with their customers, their suppliers
and their employees.
The requirement for data communications has resulted from a number of
factors. Firstly, there is a need to exchange information. For most of
mankind’s history the speed at which information could be shared over a
long distance depended on how fast a messenger could run or ride a horse.
More sophisticated methods did evolve to transmit limited information such
as the use of drums, smoke signals, beacons, semaphore signalling and
carrier pigeons, but all of these systems had their limitations. The first
recorded use of beacons is when Troy fell in about 1200 BC. Beacons are an
example of extremely simple data communications that can signal just a
single binary digit (bit) of information.
It was not until the mid-nineteenth century, with the widespread use of the
telegraph and later the telephone and radio communications, that it became
possible to communicate information between different parts of the world in
real time.
It was some years after the development of the computer that the need to
provide remote access to information arose. Initially, teletypewriter terminals
were used to access mainframe computers to carry out remote job entry by
inputting from paper tape readers and outputting to the teletype printer.
Remote Job Entry soon evolved to using Card Readers and Line Printers over
300 and later 1,200 and 2,400 bit/s modem links.
5
In the 1970s minicomputers became popular, as they did not have to be
located in large data centres, but could be sited in departmental offices under
the control of user departments rather than centralised data processing
departments. They often needed to be connected with similar machines in
other departments or with the mainframe computers. Originally they
emulated Remote Job Entry Terminals, but eventually the manufacturers
developed proprietary architectures to connect these machines. The most
common network architectures were the IBM1 Systems Network Architecture
(SNA) and the DEC2 Digital Network Architecture (DNA).
When IBM launched the Personal Computer (PC) in 1981 it had no
networking capability, but users soon discovered that they needed networks
to transfer information between machines and to access corporate data held
on mainframes. They started to connect their PC to mainframes using
terminal emulation software and modems. Enterprises also discovered that,
at the time, hard disk storage, printers and modems were very expensive if
they were provided for each PC. They could be provided much more cheaply
and efficiently as a shared resource. This led to the development of Local
Area Networks (LANs) for fast access to high performance servers for file
storage, sharing and print spooling and to modem pools. It also became more
economic to run application software on these servers.
1
International Business Machines
Inc.
2
Digital Equipment Corporation,
now part of Hewlett Packard.
The next stage was for the LANs, minicomputers and mainframes to be
linked up into an enterprise network, initially using proprietary
architectures, but eventually moving to Internet standards.
Once PCs were properly networked, other services such as email began to
develop which grew even more popular when enterprise networks began to
link up to the Internet, and users were able to send emails to other
organisations. The Internet connection also allowed access to the world wide
web (www) and other Internet services.
Networks also can be used to improve the reliability of information services.
Companies often have back-up computer facilities in a separate location to
their operational centres. Networks can be configured so that traffic can be
diverted to the back-up centre in the event of one of the operational centres
failing. Networks also allow data to be easily replicated and backed up.
Networks have made distributed processing possible. Some grid applications
such as the Search for Extra-Terrestrial Intelligence (SETI), where data is
processed by thousands of PCs across the world, would not be possible if it
were not for the ability to distribute processing via the Internet.
Networks also allow enterprises to monitor, manage and troubleshoot
remotely located equipment including the network itself. Enterprises can
save costs by centralising their technical staff.
2.1.2 Definitions
Data means distinct pieces of information3 usually formatted in a special
way. Data of itself has no meaning. It is only when it is processed by a data
processing system that it takes on meaning and becomes information.
Information is often coded as digital data expressed as a sequence of bits,
but it can also exist as continuously varying analogue data such as in human
speech and video. In this subject guide we will mainly be concerned with
digital data.
Data Communications is the transmission of data (usually assumed to be
digital data) across distances. The distances can be very short, between a
computer and a directly connected printer for instance, or can be very large,
6
3
Information was formally
defined by Claude Shannon in his
classical paper of 1948, but the
mathematical theory of
information is beyond the scope
of this course.
on a national or global scale. When the distances are large the term
telecommunications4 is often used. Telecommunications often includes the
transmission of analogue as well as digital data.
4
From the Greek word ‘tele’
meaning from far away or at a
distance.
A signal is a means of transmitting an element of data across a distance and
the process of doing this is called signalling. Signals can vary continuously
(analogue signals) or have discrete levels (digital signals). As we will see in
Chapter 8, there are four different combinations of analogue and digital data
with analogue and digital signals that can be used.
2.2 Shannon’s Communication Model
It is useful to start the study of the subject by examining a simple model of
communications. Such a model was described by Claude Shannon in his
classic 1948 paper.5 Successful communication requires the following
elements: an information source that produces information in the form of a
message; a transmitter that converts the information into signals which are
suitable for transmission through a medium (or channel) and a receiver
that receives the signal and converts it back into the format of the original
message which can be understood by the destination.
5
Shannon, Claude. A
Mathematical Theory of
Communication. (Bell System
Technical Journal, 1948).
Figure 2.1 Shannon’s Communications Model
As can be seen from the above diagram, there is a complicating factor that
affects the transmission of signals through the channel and systems. This is
noise6 which is generated by a noise source. All channels and systems are
prone to experiencing noise to a greater or lesser extent. There is no such
thing as a perfect channel or system. If a signal is amplified, then the noise
will also be amplified. Engineers spend a great deal of time and energy in
trying to design systems to minimise and overcome the effects of noise. All of
the components of the model are affected by different types of noise but the
greatest effect is upon the channel itself.
6
Noise is caused by spurious
unwanted random energy
generated inside or outside of the
channel.
7
2.3 Channels
Communications channels can be classified into a number of different types.
2.3.1 Quantitative characteristics
Channels can also be characterised by a number of different quantities:
• Bandwidth: the range of frequencies that can be effectively carried by
the channel [measured in Hertz (Hz)]. The Hertz is a measure of the
number of times a signal oscillates per second.
• Capacity: the maximum number of bits that can be carried by the
channel in one second, often also called the data rate or speed
[measured in bit/s].
• Throughput: the practical maximum number of bits that can be carried
by the channel in one second as experienced by an end user [also
measured in bit/s].
Multiplexed
FDM/TDM/STDM
Switched
Circuit Switched /
Message Switched /
Packet Switched
Point-to-Point /
Point-to-Multipoint /
Broadcast
Simplex / Half / Full
Duplex
Synchronous /
Asynchronous
Analogue /
Digital
Channel
Guided /
Unguided
Table 2.1: Types of communication channels
Telephone call
Mobile phone call
Facsimile call
TV mast to TV aerial
TV remote control to TV set
Dial-up modem to ISP
Email between servers
Web browser to server
Private circuit
Hub-based Ethernet
• Utilisation: the proportion of time that the channel is fully occupied
[measured in %].
Channels also exhibit some negative characteristics, which can be quantified:
• Noise: The causes of noise have already been discussed. Noise can be
measured by comparing its power with that of the signal using a unit
called the deciBel (dB). The deciBel is not an absolute unit. It is used to
compare the power of two signals using a logarithmic scale.
The Signal to Noise Ratio (SNR) is calculated using the formula:
SNR = 10 log (S/N), where S is the power of the signal and N is the
10
power of the noise (usually measured in Watts or milliWatts).
The following table illustrates the logarithmic relationship between the
actual ratio of signal to noise and the measurement of the Signal to Noise
Ratio in deciBels.
8
• Attenuation: the loss of power of a signal over distance [measured in
deciBels (dB)]
Table 2.2: Signal to noise ratios in deciBels
Signal : Noise
Signal to Noise Ratio (dB)
10,000 : 1
40
1,000 : 1
30
100 : 1
20
10 : 1
10
1:1
0
Here a comparison of two signal powers is made. The power of the
transmitted signal is compared with the power of the received signal. A
logarithmic unit (the deciBel) was chosen by engineers to make the
computation of cumulative losses easier. The overall loss along a circuit
can be calculated by simply adding all the individual losses on the
components that make up the circuit. Without a logarithmic scale there
would be a lot of multiplications and divisions. The deciBel was chosen
rather than the Bel7, as it allows virtually all losses and gains to be
expressed as numbers between 0 and 100. The deciBel can also be used to
calculate gains in power from amplifiers.
7
Named after
Alexander Graham Bell,
the inventor of the
telephone.
• Delay: the difference between the time a signal enters the channel at the
transmitter and the time it exits the channel at the receiver [measured in
milliseconds (ms)].
There are different components that can contribute to delay. The ones we
will be most interested in this course are:
• Propagation delay, which is caused by the limit imposed by the speed of
light on any signal being transmitted in any channel. Nothing can travel
faster than light (3 x 108 m/s) and typically communications signals
travel in wires or optical fibres travel at about two-thirds of the speed of
light (2 x 108 m/s).
• Transmission delay, which affects serial channels and results from the
time it takes to actually transmit data bits serially onto a digital channel of
a given capacity. If L bits are transmitted down a channel operating at a
speed of R bits/s, then the transmission delay will be L/R seconds.
• Queueing delay, which can affect any time division multiplexed channel
where data has to wait its turn to be transmitted. It is particularly
important with packet switched channels.
• Jitter: the variability of delay [measured in standard deviations].
Minimisation of jitter is important for many real-time applications such as
voice or video, as the presence of jitter will cause the received signal to
break up, as continuity will be lost and ultimately the signal could
become unintelligible.
Activity 2.1
In a spreadsheet, build a table to convert a Signal to Noise Ratio measured as a
straightforward ratio to a Signal to Noise Ratio measured in deciBels. You should start
with a ratio of 100:1, then 200:1 etc., up to 10,000:1.
9
2.3.2 Types of channels
Guided or unguided
In a guided channel, signals are constrained within the medium to follow a
clear path. With unguided channels there are no (or few) constraints on how
the signal propagates through the medium.8
Analogue or digital
8
Metallic or fibre optic cables are
examples of guided channels,
while radio communications is an
example of unguided channels.
Analogue channels carry analogue signals while digital channels carry
digital signals. A digital channel is therefore ideal for carrying digital
information where each bit, or sometimes a number of bits of data, can be
represented by one of a number of discrete energy values used for
transmission. In this subject guide we will mainly concentrate on digital
channels, as they offer superior performance and, as a result, modern
communications are almost exclusively digital.
Serial or parallel
In serial digital channels, data is transmitted a bit at a time, in sequence. In
parallel digital channels, a number of bits are transmitted simultaneously
over different physical paths. This subject guide will concentrate on serial
channels, as parallel channels are usually only used for short distances such
as on an internal computer bus or a printer cable.
Synchronous or asynchronous
In a synchronous channel, data arrives at fixed times and the receiver must
always be kept in step (or be synchronised) with the transmitter. The receiver
must know both the arrival time and the duration of each signal that is used
to represent the bits in a large block of data. It must be able to run a clock
that runs at the same speed as the clock in the transmitter. The two clocks
are often synchronised by transmitting a clock signal, sometimes done by
coding the data signal in a special way. In asynchronous channels, data can
arrive at any time and the receiver does not always have to be kept in
synchronisation with the transmitter. In this subject guide we will mainly
concentrate on synchronous channels, as they are necessary for high speed
communications and most channels operate synchronously today.
Simplex, half duplex or full duplex
A simplex channel only ever operates in one direction.9 A half duplex
channel operates in two directions, but only one direction at a time.10 A full
duplex channel can operate in both directions simultaneously.11 Being able
to do this is important for most applications and most modern
communications is full duplex.
Point-to-point, point-to-multipoint or broadcast
A point-to-point channel has one transmitter and one receiver. A point-tomultipoint channel has one transmitter and a specified number of
receivers.12 A broadcast channel can have one or more transmitters and any
number (often unknown) of receivers.
9
An example of a simplex
channel would be a channel used
by a TV broadcast or a stock
price feed.
10
An example of a half duplex
channel would be the channel
used by a walkie-talkie radio.
11
An example of a full duplex
channel would be the channel
used by a telephone call.
12
Point-to-multipoint channels
were used in old-style mainframe
computer networks where the
mainframe would poll a number
of devices in turn on a single
transmission line.
13
Baseband or broadband
A baseband channel is one where the channel is filled by just one signal13
while a broadband channel can carry multiple signals at the same time,
usually using Frequency Division Multiplexing.14 Note that the term
broadband can also be used to describe a high capacity channel (e.g.
ADSL15) as opposed to a narrowband channel (e.g. a telephone channel).
10
Ethernet is an example of a
baseband channel.
14
A Cable TV coaxial cable is an
example of a broadband channel.
Multiple TV channels are
multiplexed together using FDM.
15
Asymmetric Digital Subscriber
Line.
Dedicated or shared
A dedicated channel is one in which only one end-to-end communication
between different users is supported. A shared channel can support more
than one end-to-end communication. Clearly, point-to-multipoint and
broadcast channels are shared, but a point-to-point channel could either
carry one end-to-end communication or multiple end-to-end
communications. Dedicated channels are often more expensive and less
efficient than shared channels as they make less efficient use of the channel.
Shared channels can result from:
• multiplexing or switching
In order to carry more than one end-to-end communication, a channel
can allocate its capacity to a number of different end-to-end
communications simultaneously. This is known as a multiplexing
channel.16 Alternatively, a channel can allocate all its capacity to one endto-end communication at a time, and when the communication has
finished, allocate the channel to another end-to-end communication. This
is known as a switching.
Multiplexing can be achieved by Frequency Division Multiplexing (FDM),
Time Division Multiplexing (TDM) or Statistical Time Division
Multiplexing (STDM)
• With FDM, capacity on the channel is shared by allocating a different
range of frequencies to each end-to-end communication.17
• With TDM, capacity on the channel is shared by interleaving data in
individual timeslots for each end-to-end communication serially on a
round-robin basis, but each end-to-end communication is allocated its
own permanent capacity even when there is no data to be transmitted.
• STDM is similar to TDM, but there is no fixed allocation of capacity to
each end-to-end communication. They are allocated timeslots only
when needed and can, if necessary, use extra spare timeslots. STDM is
sometimes referred to as Asynchronous TDM (ATDM), as data on the
input channels can arrive at any time.
16
In practice a channel
can be both multiplexed
and switched. It can be
multiplexed into a
number of sub-channels
each of which could
then be switched. An
example of this is the 2
Mbit/s circuit used by
network operators and
enterprises to carry
voice traffic. Each
2Mbit/s circuit is
permanently divided
into thirty 64 kbit/s
voice channels onto
which telephone calls
are switched.
17
For high-speed
optical fibres, the term
used for FDM is
Wavelength Division
Multiplexing (WDM).
Switching can be achieved by circuit switching, message switching or packet
switching.
With circuit switching, an end-to-end communication seizes all the capacity
of the channel for a fixed period of time by setting up a connection through
the channel. With circuit switching, the channel can be idle while no data is
being transmitted and the channel cannot be used by other end-to-end
communications until the connection is released.18 Circuit switching
therefore does not usually make the most efficient use of the channel.
With message switching, an end-to-end communication consists of a single
arbitrarily long message, which is transmitted through the channel, but as
soon as the message has been transmitted, the channel is free to transmit
another message from another end-to-end communication.19
Packet switching is similar to message switching except that packets cannot
be of arbitrary length. Instead, messages are broken up into packets which all
have a maximum size. Packets from different end-to-end communications
can be interleaved on the channel.20
With message and packet switching the capacity of the channel is allocated
much more dynamically than in circuit switching, and the channel is only
idle when there are no end-to-end communications transmitting. The whole
capacity of the channel can be allocated to just one end-to-end
18
A good example of
circuit switching is a
64kbit/s path between
two telephone
exchanges that can be
allocated to any
telephone call.
19
E-mail as an
application is an
example of messageswitching between mail
servers.
20
The Internet uses
packet switching.
11
communication, if it is the only one that requires capacity. Hence they are
much more efficient, but delays also become variable and in the case of
message switching can be very long.
Packet switching can be further subdivided into two distinct types of
switching:
• Virtual circuit based (or connection-oriented switching) where a
virtual circuit is set up through a network and all packets follow the
same path through the network.21 The effect of this is in some ways
similar to circuit switching, but a virtual circuit uses no pre-allocated
network capacity and hence is more efficient than circuit switching.
21
Asynchronous Transfer Mode
(ATM) is an example of a virtual
circuit switched network.
Virtual circuits can be switched or permanent. Switched virtual
circuits are set up as and when needed by means of a special call
request packet which is allocated a virtual circuit number. This
virtual circuit number is used in subsequent packets (instead of an
address) to ensure that they follow the same route. When data
transfer is complete the virtual circuit is disconnected and the virtual
circuit number can be reused by other users. Permanent virtual
circuits also use virtual circuit numbers but they are permanently set
up by network administrators.
• Datagram based (or connectionless switching) where each packet is
routed independently based on an address that the packet contains
and each packet can follow a different path through the network.22
A channel can also be composite, built from a number of different
components and media in series to provide a communications path between
the transmitter and the receiver.23
Activity 2.2
Consider different types of communication channels used in the various types of
communication in Table 2.1 and any others you can think of. Attempt to classify them by
type in the table. The last two columns only relate to channels that are shared in some
way.
22
The Internet is an example of a
datagram network.
23
A good example of a
composite channel would be a
telephone channel, which will
typically pass through many
switches, multiplexers and copper
and fibre optic communications
circuits but will be able to carry a
voice signal from the transmitting
phone to the receiving phone.
2.4 Networks
2.4.1 Definitions
A network is a collection of systems that are interconnected to exchange
information with one another and share resources.
A computer network is a network that comprises terminals, computers,
servers and other components (usually owned and managed together).
A host is a computer that runs (or hosts) end-user applications.
An internetwork or internet24 is a collection of interconnected computer
networks. Sometimes these networks are incompatible (running different
protocols25 and addressing schemes) and must be interconnected by
Gateways, which translate between the different protocols. A subnetwork26
(or subnet) is the part of a computer network left after all the hosts have
been excluded.
• A Wide Area Network (WAN) is a computer network that covers a
country or a continent or the whole world.
• A Metropolitan Area Network (MAN) is a computer network that covers
a campus or a city.
• A Local Area Network (LAN) is a computer network that covers a limited
geographical area (usually a single building or a part of a building).
12
24
The word internetwork or
internet is a generic term and the
global Internet (with a capital I) is
a specific example of an
internetwork.
25
Protocols are sets of rules
governing the procedures and
data formats used for
communication.
26
Subnetwork also has a more
specialised meaning in the
Internet. Here a subnetwork (or
subnet) can mean an individual
network (such as an Ethernet)
with a separate addressing range
that forms part of an
internetwork (such as the
Internet).
• A Personal Area Network (PAN) is a new type of computer network
that is used to connect devices such as mobile phones, personal
computers, personal digital assistants (PDAs) and peripherals, close to
one person.
Large networks are also sometimes designed in a hierarchical structure with
three clearly identifiable layers:
An access network is a network that supports end users. Access networks
can be residential (e.g. local telephone, cable TV and ISP networks),
institutional (e.g. a University or an office) or mobile where users are free to
roam (e.g. a wireless LAN). Access networks are not normally resilient.27
A distribution network does not support end users but concentrates traffic
from access networks and forwards it to a core network. Distribution
networks often provide partial resilience.
27
Resilient means
being able to withstand
failures.
A core network is a network that can switch or route large volumes of traffic.
A failure of a core network will affect all users and hence it has to be highly
resilient and stable.
2.4.2 Network topologies
Topology is a branch of Mathematics which is concerned with how things are
connected. It takes no account of distances or spatial positions. It has been
described as ‘rubber band geometry’. Topology is highly relevant to the study
of networks. Networks can be classified according to their topologies. There
are a number of basic network topologies:
• Bus networks (Figure 2.2) are often used in LANs, but not MANs or
WANs. All the nodes share access to a common medium. Buses are easy to
implement but are not resilient to failures.28 A cable fault will split the
network in two. Isolating a fault on a bus is also very difficult but adding a
new node is easy.
• Ring networks (Figure 2.3) are used for LANs, MANs and WANs.29 They
can provide resilience to single failures, if data can travel around the ring
in either direction. Ring networks are harder to implement than bus
networks and use more cable, but faults are much easier to isolate.
Adding a new node is difficult.
• Star networks (Figure 2.4) are used in both LANs and WANs.30 They are
not particularly resilient to failures, especially at the hub or central site. It
is relatively easy to add new nodes, and star networks are easy to fault
and manage as most of the network equipment is centralised.
• Tree networks (Figure 2.5) are really extended buses; they could actually
be described as Bus-Bus31 hybrid networks where there is a single
medium which branches in various places (not at nodes).32 They possess
little resilience and fault isolation is difficult, but new nodes can easily be
added.
• Mesh networks are normally used in WANs and they are relatively easy
to manage and fault. They come in two varieties:
• Fully-meshed networks (Figure 2.6) are usually only economic for
small networks or in core networks. They are extremely resilient to
multiple failures. They are very expensive if there are a large
number of nodes. Adding a new node is difficult and expensive.
28
The original thick
and thin Ethernets use
bus topologies. All the
stations were
connected into a
coaxial cable which was
run around a building.
29
IBM Token Ring
LANs and modern wide
area transmission
systems such as
Synchronous Digital
Hierarchy (SDH) use
ring topologies.
30
Twisted pair
Ethernets and
mainframe computer
networks, where
remote terminals access
a central site, use star
topologies.
31
See terminology used
to describe hybrid
networks.
32
Bridged Ethernet
LANs and legacy
mainframe computer
networks use tree
topologies; the latter
via multidrop (point to
multipoint) circuits.
• Partially-meshed networks (Figure 2.7) are resilient to multiple
failures depending on the degree of connectivity. They are not as
expensive as fully meshed networks. Adding a new node is
somewhat easier and less expensive than for a fully-meshed
network.
13
• Hybrid networks are composed of two or more of the basic types and
inherit the advantages and disadvantages of their component topologies.
They are often described by hyphenating the basic types, with the second
term indicating the topology at the centre of the network:
• Star-Bus networks (Figure 2.8) have a central bus which connects
together a number of star networks.33
• Star-Star networks (Figure 2.9) are commonly known as Cascaded
Star Networks as there is one central star network to which other star
networks are connected.34 This topology has little resilience,
particularly if all traffic has to pass through the central node.
• Star-Ring networks (Figure 2.10) have a central ring which connects
a number of star networks.35 It has a resilient core but does not
provide resilience to each outlying node.
• Star-Mesh networks (Figure 2.11) have a central mesh network to
which a number of star networks are connected.36 It has a very
resilient core but does not provide resilience to each outlying node.
• Ring-Ring networks (Figure 2.12) are also known as Multiple
Overlapping Rings (or Shared Protection Rings or SPRings). The
topology is a partial mesh network, but is constructed in a special way.
It comprises a set of rings with shared segments.37 It provides a high
degree of resilience to multiple failures at low cost.
14
Figure 2.2: Bus Topology
Figure 2.3: Ring Topology
Figure 2.4 Star Topology
Figure 2.5 Tree Topology
33
Star-Bus networks are
sometimes used in LANs where
each floor has a star-connected
Ethernet and the hub on each
floor is connected to a Fast
Ethernet bus, which connects all
the floors.
34
Star-Star networks are used in
old-style mainframe WAN
networks where cluster
controllers were positioned in
such a way as to support remote
terminals with minimum circuit
costs.
35
Star-Ring networks are often
used in LANs where each floor
has a star-connected Ethernet
and the hub on each floor is
connected to a ring, which
connects all the floors.
36
Star-Mesh networks are used
in large WANs where there is a
partially (or fully) meshed core
connected to distribution
networks in a star configuration.
37
SPRings have become
increasingly popular with the
telecommunications operators
who use them for their
Synchronous Digital Hierarchy
transmission networks.
Figure 2.6 Full Mesh Topology
Figure 2.7 Partial Mesh Topology
Figure 2.8: Star-Bus Topology
Figure 2.9 Star-Star Topology
Figure 2.10: Star-Ring Topology Figure 2.11: Star-Mesh Topology
Figure 2.12: Ring-Ring networks
15
Activity 2.3
Investigate the topologies of the local area and wide area networks that are deployed by
your educational institution, business or other enterprise in your locality. You may need
to talk to a technician or ask to see some network maps. Which network topologies are
employed and why? How resilient are the topologies to failure?
Activity 2.4
Explore the Atlas of Cyberspace web site (http://www.cybergeography.org/atlas) and
examine and attempt to classify the topologies used by Internet Service Providers.
a. State whether each of the following statements is true or false and, if false,
correct the statement:
i. Noise affects all the components of Shannon’s Communications Model
and not just the channel.
ii. All packets using virtual circuits contain a full network address.
iii. Propagation delay results from the time it takes to actually transmit data
bits serially onto a digital channel.
iv. A sub-network is the part of computer network left after all the hosts
have been excluded.
b. Describe the main differences between synchronous and asynchronous channels.
c. Explain why jitter is undesirable for real-time communications such as voice and
video communications.
d. A channel experiences an average noise power of 0.2 mW and the average
power of the signals it carries is 200 mW. What is the Signal to Noise Ratio in
deciBels? If the signal is amplified by 40 dB, what will be the average power of
the noise?
e. For a WAN consisting of 6 nodes, draw the topology with the least number of
point-to-point links and draw a topology that best improves the resilience of this
network by adding a single link. Draw another topology that will maximise
resilience for the whole network and another that will minimise delays between
one node and all of the others.
Learning outcomes
At the end of this chapter and the relevant readings, you should be able to:
• outline the history of data communications
• describe Shannon’s Communication Model and all its components, including the
various sources of noise which affect all communications channels
• explain the qualitative and quantitative characteristics of a channel
• classify communication channels according to their characteristic type
• calculate gains, losses and Signal to Noise Ratios in deciBels
• classify networks according to their type
• identify different basic and hybrid network topologies and list their advantages
and disadvantages.
16
Introduction
In this chapter, we will examine what is meant by layered network
architectures and why they are needed. We will study the different types of
standards and the organisations that produce them. We shall consider how
network devices can be classified in relation to the hybrid reference model
and to their role within the network. Finally, we shall examine the concept of
a protocol and introduce some general terminology that can be used in
describing the operation of protocols.
Further reading
third edition, (2003). Chapters 1.4, 2.1, 2.2, App 3.
Kurose, James F. and Keith W. Ross Computer Networking. (Addison-Wesley),
third edition, (2005). Chapters 1.7.1.
International), seventh edition, (2003). Chapters 0.3, 2.1.2, 2.3, 2.4, 18.1.
fourth edition, (2002). Chapters 1.3, 1.4, 1.6, 4.7.5.
3.1 Layered architectures
When building complex structures or systems, engineers like to define an
architecture. Computer hardware, software and networks are easier to
design, build, test and maintain if they are based on a well defined
framework or architecture. A network architecture can be defined as a
highly structured framework within which networks can be analysed,
designed and implemented, incorporating a defined set of layers and
protocols.
Computer and network architectures are defined as a set of layers which
perform different functions and can be implemented separately with well
defined interfaces between each layer. The lower layer provides a service to
the layer immediately above it by means of an interface. Think of a layer as
software implementations that perform specific functions and the interface
as a set of procedure calls with parameters. With network architectures there
must be two similar implementations at both ends of a communications
channel. These layers are called peers and they communicate with each
other using a protocol (or set of rules) specific to that layer. The set of layers
is often referred to as a protocol stack or protocol suite.
With layered architectures:
• each layer is functionally independent
• each layer has a well defined interface to the previous layer
• each layer communicates indirectly with its peer layer at the opposite end
of a communications channel using a protocol specific to that layer
• each layer communicates with its peer layer via an interface with the
layer immediately below, unless it is the bottom layer, in which case it
communicates directly with its peer.
The advantages of layered architectures are as follows:
17
• they divide complex operations into more manageable groups which are
then more easily implemented and tested
• it is possible to change one layer without affecting all the others,
providing that the same interfaces are supported1
• it is possible to mix and match different technologies and suppliers for
different layers if they support standard interfaces.2
The advantages of layered architectures are clear; the only question is how
many layers are needed and what functions are to be placed in each layer.
There are two main models for defining network architecture. The
International Organization for Standardization (ISO) has defined a seven
layer model called the Open Systems Interconnection (OSI) Reference
Model. This model was developed in the 1980s as part of an exercise to
produce open standards to free users from the proprietary standards which
were being used by computer vendors to lock customers into their network
architectures. At the time the vast majority of customers were using IBM’s
proprietary architecture called Systems Network Architecture (SNA) which
also had seven layers. OSI was not particularly successful. Both OSI and SNA
were soon overtaken by the development of the Internet. The researchers
working for the US Military who developed the Internet used their own four
layer model called the Internet or Department of Defense (DoD) Model. This
model is also sometimes called the TCP/IP Reference Model, named after the
two main Internet protocols.3 Tanenbaum gives a good comparison of the
two models and an interesting insight into the reasons why the Internet
protocols now predominate.4
Both ends of a communications channel must implement the same standards
for each layer in order to communicate. These standards are called protocol
standards. They define the syntax of the data to be transmitted at this layer
and all of the other rules which govern communications between peer
entities at this layer. Communication between the layers at both ends of the
channel is not direct (apart from at the very bottom layer). Instead each layer
relies on the layers below it to effect communications. It makes use of the
services provided by the layer immediately below which may in turn rely on
services provided by another layer below. In effect, what actually happens is
that each layer adds a protocol header to the data that is to be sent. A
protocol header contains information that is only relevant to the peer layer at
the other end of the channel. When a packet of data is received by the peer
layer at the other end of the channel, the protocol header is processed and
then stripped away before the data is passed to the layer above. In this way
the upper layers are not aware of or do not need to know anything about the
protocols or headers being used by lower layers.
Before we look at network architecture in detail, it will be useful to define
some commonly used terms. An interface exists between two adjacent layers
through which the lower layer provides a service to the upper layer. The
interface is formally specified by a set of primitives which are implemented
as procedure calls with parameters and which are executed whenever the
upper layer requests a service from the lower layer.5 The Service Access
Point (SAP) is the place at which the service is provided and is identified by
a unique Service Access Point (SAP) Address.6
18
1
For example, a carrier would
upgrade its network layer from IP
version 4 to IP version 6 without
having to make any changes in
the transport layer above or the
data link layer below.
2
For example, a Local Area
Network could be changed from
an IBM Token Ring to an
Ethernet at one site without
affecting the Internet Protocol
and application layer software
that generates and receives data,
or without requiring a change to
the LAN used at any other site
with which the first LAN
communicates.
3
Transmission Control protocol /
Internet Protocol.
4
Tanenbaum, Chapter 1.4.3,
p. 44.
5
E.g. The IP network service has
just two primitives: SEND to send
a packet throughout the network
layer and DELIVER to receive a
packet from the network layer.
6
E.g. The Network Service
Access Point for the Internet
Protocol is identified by a unique
Network Service Access Point
(NSAP) address. This NSAP
address for IP is more commonly
known as the IP address.
3.2 Reference models
You studied the Open Systems Interconnection (OSI), Internet and Hybrid
Reference Models in a Level 1 unit. You should re-familiarise yourself with
the functions of the layers in each model. The names of the layers are
summarised in Table 3.1, below.
Table 3.1: Comparison of OSI and Internet reference models
This course will assume the hybrid model which attempts to make use of the
best features of both models to produce a practical and simple model that
can be used to describe network architectures. This model has five layers and
its relationship with the OSI and Internet models can be seen in Table 3.1.
Activity 3.1
Look at the web site http://www.protocols.com/protocols.htm. You will see a table with a
long list of protocols in the first column. Find some protocols that you have heard about
and click on the link to the family of protocols that these protocols belong to. If you have
not heard of any of the protocols then click on some of the families of protocols in the
third column of the table. For most families, by scrolling down the page, you will be able
to see a table that shows how the protocols in that family relate to the OSI Reference
Model.
3.3 Network standards
One word that has occurred a number of times so far in this section is
‘standard’. Standards allow different vendor products to inter-work. They can
be de jure (by law) if they have been produced or accepted by a recognised
standards body or de facto (by fact) if they have not been produced or
accepted by a recognised standards body, but have gained widespread use by
market forces. Hardware and communications standards are often de jure
while software standards are often de facto. A good example of a de facto
standard is the Microsoft Windows operating system. Competitive markets
will only usually arise after de jure standards have been agreed and have
stabilised. Network vendors work together in standards bodies to agree
standards for new technologies. De facto standards can be further divided into
proprietary (or closed) standards which are produced, owned and controlled
by a commercial organisation (e.g. Microsoft Windows) and open standards
which, while they may have been originally produced by a commercial
organisation, have since been transferred to the public domain (e.g. Unix).
19
There are a large number of standards bodies which develop standards for
data communications and networks. The main international standards
bodies are:
• International Organization for Standardization (ISO) whose members are
various national standards bodies such as the American National
Standards Institute (ANSI) in the USA, the British Standards Institute
(BSI) in the UK, Association Française de Normalisation (AFNOR) in
France, and Deutsches Institut für Normung (DIN) in Germany.
ISO produces a wide range of standards in many areas other than
telecommunications. Its standards are prefixed by ISO.7
• International Telecommunications Union Telecommunications
Standardization Sector (ITU-T) whose members are national
governments, regulatory agencies and various equipment suppliers and
network operators. It is a part of the United Nations and its standards,
called recommendations, are prefixed by a letter of the alphabet.8
• European Technical Standards Institute (ETSI) whose members are
administrators, manufacturers, network operators, service providers and
users, in Europe and other countries. It produced the GSM standard for
second generation mobile phones. Its telecommunications standards are
prefixed by ETSI EN.9
In addition to these international standards bodies, there are a number of
other bodies which set standards in specific areas of networking. These
include professional bodies and forums from industry and academia
interested in developing new standards. Examples of these bodies are:
• Institute of Electrical and Electronic Engineers (IEEE) which has been
particularly active in the standardisation of LANs. Its standards are
prefixed by IEEE.10
• Internet Engineering Task Force (IETF), which is responsible for all
standards relating to the Internet. Its standards are prefixed by RFC which
stands for Request For Comment which continues to be used after a
standard has been agreed.11
• Electronic Industries Association / Telecommunications Industry
Association are two American trade associations which have been
accredited by ANSI to develop standards for the electronic and
telecommunications industries. They are both involved in developing
computer/network interface standards. Their standards are prefixed by
EIA/TIA.12
• Frame Relay Forum which is an industry group which was set up to
standardise Frame Relay protocols. Standards, called implementation
agreements, are referenced by numbers prefixed by FRF.13
• ATM Forum which is an industry group which was set up to standardise
Asynchronous Transfer Mode protocols. ATM standards are referenced by
numbers prefixed with af.14
Activity 3.2
Use a search engine to find the websites of the standards bodies referred to in this
chapter, and explore these sites to see what standards work they have been engaged in.
3.4 Network devices
Network devices can be classified according to the highest layer at which they
operate. This layer is the highest layer for which they need to be able to read
the protocol headers. All network devices will be able to forward application
20
7
For example, ISO 4335.
8
For example, G.723, Q.931,
V.24, X.25.
9
For example, ETSI EN 301
419–1.
10
For example, IEEE 802.3.
11
For example, RFC 791.
12
For example, EIA/TIA 232F.
13
For example, FRF 1. 2.
14
For example, af-phy-0015.
000.
layer data, but they do not have to understand what the application layer
headers and data mean. A device is therefore regarded as a network layer
device if it has software that enables it to read network layer headers, but not
transport layer headers, in order to carry out its function. A network layer
device will, of course, require software to handle the data link and physical
layers.15
The hosts will always require a full protocol stack and will need software to
read application headers and data as well as all the other headers in the
protocol stack (See Fig 3.1). Other devices may only need to be able to read
headers from network layer and below (See Fig 3.2), and others will only
need to read data link headers (See Fig 3.3). Some devices will not even
need to be able to read the data link headers. These will be physical layer
devices (See Fig 3.4).
Activity 3.3
Explore the RFC Search web site (http://www.rfc-editor.org/rfcsearch.html) and perform
an RFC search for some protocols that are used on the Internet. Examine a few of the
RFCs that specify these protocols to see how they are written.
Figure 3.1: An application layer intermediate device between two hosts
Application
Application
Application
Transport
Transport
Transport
Network
Data Link
Network
Data Link
Network
Data Link
Physical
Physical
Physical
15
As an example an IP
Router is a network
layer device, as it has
software that enables it
to read the IP address
in an IP header in order
to route an IP packet.
But it also has software
on its line cards to
handle Ethernet and
serial ports running
different data link
protocols. It may
receive a frame from an
Ethernet, read the data
link address to check
that it is the correct
destination for the
frame, then read the IP
header to find the
destination of the
packet and look this up
in its routing table to
discover which port
should route the
packet. Finally, the line
card which supports
this port has to
encapsulate the IP
packet in the data link
protocol appropriate to
that port. This process
is shown conceptually
in Figure 3.2.
Figure 3.2: A network layer intermediate device between two hosts
Application
Application
Transport
Transport
Network
Data Link
Network
Data Link
Network
Data Link
Physical
Physical
Physical
Figure 3.3: A data link layer intermediate device between two hosts
Application
Application
Transport
Transport
Network
Data Link
Data Link
Network
Data Link
Physical
Physical
Physical
21
Figure 3.4: A physical layer intermediate device between two hosts
Application
Application
Transport
Transport
Network
Data Link
Network
Data Link
Physical
Physical
Physical
It should be noted that because transport protocols operate end-to-end there
are very few requirements for intermediate transport layer devices. There are
however many intermediate application layer devices in common use.16
Another classification of network devices is whether they are DTEs or DCEs.
DTE stands for Data Terminal Equipment and DCE can either stand for
Data Communications Equipment (according to the EIA) or Data Circuitterminating Equipment (according to the ITU). The distinction originates
from the need to have different interface specifications at either end of the
cable between a data terminal (such as a PC) and a modem.17 The PC is
configured as a DTE and the modem as a DCE. Another important difference
between the two occurs when the modem supports synchronous
communications; the modem must provide a timing signal to the PC to
maintain synchronisation. Another key distinction is that the interface on a
DTE is normally male and the cable connector that plugs into it must be
female. DCEs normally have a female interface and the cable connector that
plugs into them must be male.
As a general rule, equipment that historically belonged to or still belongs to
network operators is usually configured as a DCE and equipment belonging
to their customers (that is a source or destination of data) is configured as a
DTE. Most modems are now owned by customers, but they must still be
configured as DCEs. You may think that this distinction is not very important,
but when you try to connect two different local devices with a cable, it is
crucial. If one device is a DCE and one is a DTE, then they can be connected
by a straight-through male-to-female cable where each pin on one connector
is connected via the cable to an equivalent pin on the other connector.
However, if both devices are DTEs18 or both devices are DCEs, this straightthrough cable will not work and a cross-over cable is required, where some
pins on one connector are not connected to the equivalent pins of the other
connector. On many network devices, such as routers, the ports can be
configured by software either to be DTEs or DCEs.
Activity 3.4
Using textbook indexes or web searches, read about the network devices in Table 3.2
and complete the table by determining the layer of the hybrid model at which they
operate and whether they are normally configured as DTEs or DCEs.
22
16
A good example of an
intermediate application layer
device, is a mail relay that reads
e-mail headers and decides
where the message should be
forwarded.
17
On a standard EIA232
interface pin 2 is called transmit
and pin 3 is called receive, but
only one end of the interface
cable can transmit on a pin 2.
The other end must receive on
pin 2. Likewise with pin 3. If both
ends were configured in the same
way, communication would be
impossible. In this case, the
device that transmits on pin 2 is
the DTE and the device that
transmits on pin 3 is called the
DCE.
18
If you try to connect up two
PCs back to back using a cable, a
female-to-female cross-over cable
is required as both will be
configured as DTEs.
Table 3.2: Network devices
Device
Layer
DTE/DCE
Amplifier
Asynchronous Transfer Mode (ATM) Switch
Bridge
Digital Subscriber Line Access Multiplexer (DSLAM)
Frame Relay Access Device (FRAD)
Front End Processors (FEPs)
Gateway
Hub
Inverse Multiplexer
Local Area Network (LAN) Switch
Modulator/Demodulator (Modem)
Multiplexer
Network Terminating Equipment (NTE)
Repeater
Router
Packet Assemblers/Disassembler (PAD)
Packet Switch
Private Automatic Branch Exchange (PABX)
Telephone Switch
3.5 Network protocols
There are also specialised names used to describe data packets at various
points in the layered network architecture. When layer N+1 wishes to
communicate with its peer layer, it does so by transferring a packet known as
a layer N+1 Service Data Unit (SDU) across the interface with the layer N
immediately below, along with some control information. Layer N then adds
some protocol header to the Layer N+1 SDU to form a larger packet called a
layer N Protocol Data Unit (PDU), which consists of the layer N header
consisting of a number of different fields used by the protocol, followed by the
layer N data field19 which is actually the layer N+1 SDU. The layer N header
contains information only relevant to layer N and it is sent just ahead of the
19
The data field is also
often referred to as the
payload.
23
data. The PDU is transferred to the peer layer N entity, possibly via lower layer
SDUs and PDUs. The peer layer N entity then strips off and processes the layer
N protocol header and passes the SDU up to layer N+1.
Figure: 3.5: SDUs and PDUs
Layer N Header
Layer N + 1 Data
Layer N + 1 SDU
Layer N + 1 Header
Layer N + 1 Data
Layer N + 1 PDU/ Layer N SDU
Layer N + 1 Header
Layer N + 1 Data
Layer N PDU / Layer N - 1 SDU
Instead of using numbers to describe layers, names and letters are often
used. For instance, in OSI terminology, a Network Service Data Unit (NSDU)
will be passed from the Transport Layer to the Network Layer via a Network
Service Access Point (NSAP) which will add appropriate Network Layer
headers to it to create a Network Protocol Data Unit (NPDU). This will be
forwarded to the peer Network Layer, which will then strip off the Network
Layer headers and pass the NSDU up to its Transport Layer.20
In this subject guide we will use the more common terminology and use the
word message to describe an Application SDU/PDU; segment to describe a
Transport SDU/PDU; packet to describe a Network SDU/PDU; and frame to
describe a Data Link SDU/PDU.
The interfaces between layers (Service Access Points in OSI terminology) are
implemented as a set of primitives. These can be thought of as a set of
software procedures or functions with parameters. One of these parameters
will be the Service Data Unit to be transmitted, or more accurately a pointer
to a memory buffer that contains the SDU. OSI primitives are of four basic
types as follows. A Request is issued from a higher layer to a lower layer to
request a service, which is then carried by a PDU to the peer layer
whereupon it is passed up to the higher layer as an Indication. The higher
layer at the distant end then issues a Response to the Request to its lower
layer which is again transferred via a PDU to the requester where it is passed
back up to the higher layer as a Confirmation.
Network architectures are defined by a set of protocol standards for each
layer in the reference model. The standards are developed and agreed by
various standard bodies.
A protocol is a set of mutually agreed rules that allows two peer layer
entities to communicate with each other.
Protocols can be as follows.
3.5.1 Symmetric/asymmetric
Some protocols are completely symmetric. The implementation at both ends
of the communications channel is identical and all the functions provided by
the protocol can be invoked from either end. These protocols are often
referred to as peer-to-peer. Other protocols are asymmetric where one end
can invoke functions that the other end cannot, in which case, the protocol is
often referred to as a master/slave or a client/server protocol.
3.5.2 Standard/proprietary
Protocols can either be standard or proprietary. Standard protocols are
controlled by official standards bodies, whereas proprietary protocols are
controlled by commercial organisations. Customers today are wary of
proprietary protocols, as their use tends to lock them into suppliers.
24
20
Or even more simply using
TCP/IP terminology; a TCP
segment is passed from the TCP
layer to the IP layer via the IP
interface, which adds an IP
header to it and then transmits it
as an IP datagram.
3.5.3 Connection-oriented/connectionless
This is an important distinction. A connection-oriented protocol requires
that a connection is set up between both ends before any data can be
transmitted. With a connectionless protocol, data can be sent at any time
without any need to set up a connection beforehand. With a connectionoriented protocol some state information regarding the connection will
be required at all the nodes involved in the connection, and if this state
information is lost due to a node crashing the connection will disappear
and will have to be re-established. Connectionless protocols do not suffer
from this problem and are hence more resilient. Because of this
connection-oriented protocols are sometimes called stateful protocols
and connectionless protocols are often called stateless protocols.
Connection-oriented protocols do, however, have some advantages. They
can be more efficient, as protocol headers can be shorter, because each
data packet does not have to be individually addressed. Also because
there is a connection set-up, it is possible to allocate resources to support
the connection at every node and hence the quality of service can be
guaranteed. In other words, all data packets can be guaranteed to arrive
on time, in order and without loss or duplication. It is not possible to
guarantee quality of service with connectionless protocols.
Protocols in different layers of a network architecture will have many
different functions, but a large number of protocol functions appear in
similar forms in many of the layers. They are based on the generic protocol
functions listed by Stallings21 and are listed below:
21
Stallings,
pp. 575–580.
3.5.4 Encapsulation/de-capsulation
This function has already been discussed, although the terminology is new. A
layer receives an SDU from the layer immediately above, adds its own
protocol header to the SDU, which is encapsulated in the PDU and then
transmitted to the peer layer, which strips off the header and passes the decapsulated SDU to the layer immediately above.
3.5.5 Segmentation/reassembly
It is sometimes the case that the SDU received from the layer
immediately above is too big to be carried in a PDU. In this case the layer
must segment the SDU into one or more PDUs and send these separately.
The peer layer must recognise such segmented PDUs and reassemble
them into the original size SDU before passing them up to the layer
immediately above. In the case of IP, this process is called fragmentation
rather than segmentation, as TCP uses the term segment to describe its
PDUs. In TCP/IP, a TCP segment that is too large for an IP network layer
to transmit in one packet is fragmented into one or more smaller packets
and is reassembled by the peer network layer. Protocols which support
segmentation will contain some kind of sequence number field in their
headers to assist in reassembly.
3.5.6 Multiplexing/de-multiplexing
One layer can often carry many SDUs between several different pairs of
communications entities in the layer above to make efficient use of lower
level services. This is called multiplexing. The PDUs must contain
information (usually addresses) that identifies the entities involved so that
the peer layer can de-multiplex PDUs to the correct entity.
25
3.5.7 Addressing
In order to communicate with an entity at any layer, it is necessary that the
entity can be uniquely identified. Most PDU headers contain an address field
which uniquely identifies the peer entity to which it is communicating.3.5.8
3.5.8 Ordered delivery
With connectionless protocols, PDUs can arrive out of order as they may take
different routes. When this happens the layer must buffer and reorder the
PDUs higher.
3.5.9 Priority
It is sometimes necessary for certain PDUs to be given preferential treatment
in queues so they can be transmitted with minimum delay. This may be
determined by a higher layer, such as an application that operates in real
time and thus has a requirement for low delays. It may also be used to allow
certain higher level commands, such as those that will interrupt a process on
the destination to overtake some of the data that would otherwise arrive
beforehand and be processed. To indicate the priority of a PDU, there is often
a field in the PDU header that will indicate its priority from a number of
different levels.
3.5.10 Error control
PDUs can be corrupted or lost in transmission. Fields in PDU headers such as
sequence numbers or error detection codes can be used to detect errors.
Once errors are detected, some protocols will attempt to recover from the
errors. Other protocols (especially connectionless ones) will simply discard
PDUs in which errors are detected.
3.5.11 Flow control
Sometimes a transmitter can send PDUs at a faster rate than a receiver (or
even the network itself) can process them. In such situations it is useful to
have a flow control mechanism to regulate the speed of transmission. This is
often achieved by means of a procedure whereby PDUs referenced by a
sequence number in the protocol header are acknowledged. Connectionless
protocols usually do not support flow control, but will simply discard PDUs if
they are arriving at too fast a rate to be processed.
3.5.12 Connection control
For connection-oriented protocols, a mechanism must exist to establish and
close a connection. This is usually achieved by means of special PDUs, whose
meaning is defined in a type field in the protocol header, to request the
establishment of a connection or to request an existing connection be closed.
In order to ensure a consistent state at both ends of the connection, the
requests will have to be acknowledged by another special PDU indicated by
its type field.
3.5.13 Grade of service
When establishing connections, it is useful to specify what performance is
required from the connection in terms of such parameters as delay, jitter or
throughput. These parameters can define the grade of service required for
the connection. The PDU that establishes a connection will often contain a
field to specify the grade of service. If the receiver and the network can
26
support the grade of service requested, the PDU will be positively
acknowledged. If the grade of service cannot be supported the required
grade of service parameters can be negotiated downwards or the connection
could be immediately closed.
3.5.14 Security
A number of protocols contain built-in security features to perform functions
such as authenticating the parties involved in the communication. This is
also achieved by means of authentication fields in the PDU headers.
3.5.15 Data encoding
Protocols must define the syntax (format) of data fields in PDUs as well as
the semantics (meaning). The rules for how data in a PDU is represented
have to include such things as the type of the data22 and the length of the
field in bits or bytes. The semantics of the data is defined in rules that ascribe
meaning to each allowable value of the data field.
22
For example, integer,
character string, etc.
3.5.16 Data encryption
Data encryption is closely associated with security, but can relate to either
the whole PDU or just the data field of the PDU, which needs to be scrambled
so that parties other than the recipient (who alone has the facility to decrypt
the data) cannot view the data being sent.
3.5.17 Data compression
Transmitting some types of data can be very inefficient. Transmitting bit
mapped images, full video and audio all require large volumes of data to be
transmitted and hence will require either large transmission capacity or a
long transmission time. However, these types of data contain a large amount
of redundant information that can be compressed (often without a
perceptible loss of quality) so that it can be transmitted more efficiently.
Some protocols provide a capability to compress data.
i. The application layer in the hybrid model corresponds to the top two
layers of the OSI Reference Model.
ii. ITU-T standards are called recommendations.
iii. When connecting a PC to a modem, the PC has a DTE interface and the
modem a DCE interface.
iv. Protocol Data Units are passed across the interface between adjacent
layers in a host.
b. What is meant by a layered architecture and what advantage is obtained from
using such an architecture?
27
c. List the full expanded names of three standards bodies and give an example of a
standard that each body has defined.
d. Describe the main differences between connection-oriented and connectionless
protocols.
e. Describe how protocols are encapsulated and decapsulated.
Learning outcomes
At the end of this chapter and the relevant reading, you should be able to:
• list the characteristics and advantages of layered architectures
• identify the main standards bodies involved in the development of protocols and
which types of protocols they specify
• distinguish between de jure and de facto standards
• identify which network devices operate at which layer of the Hybrid Reference
Model, and which are normally configured as DTEs and which as DCEs
• distinguish between the characteristics of a DTE and a DCE
• describe how data is transmitted within a layered architecture using correct
terminologies
• distinguish between standard and proprietary protocols, symmetric and
asymmetric protocols and connection-oriented and connectionless protocols
• describe the generic functions of protocols which can be applied to protocols in
any layer.
28
Introduction
In this chapter, we shall examine the services and interfaces provided by the
application layer and the general functions of application layer protocols. We
shall then examine in some detail virtual terminal, web, email, file transfer
and access, directory and network management protocols.
A very large number of application layer protocols have been designed for
many different applications. We will only examine in detail a small number
of protocols that are commonly used on the Internet, plus a few ISO
protocols in outline.
This subject guide will not cover all the details of protocols (either the fields
in the protocol headers or the protocol mechanisms). Rather it will attempt
to summarise their basic functions. You are advised to study in detail the
main protocols used on the Internet. Details of these protocols can be found
in any networking textbook, or from looking up the RFCs, or from websites.
Further reading
third edition, (2003). Part 6, Chapters 24.1, 25, 26, 27.
Kurose and Ross, Computer Networking – A Top Down Approach featuring the
Internet (Addison Wesley), third edition (2005). Chapter 2.1.1, 2.2.1–6, 2.3,
2.4.
Stallings, William, Data and Computer Communications. (Prentice Hall
International), seventh edition (2003). Chapter 22.
fourth edition, (2002). Chapters 5.4.1, 7.1–7.3.
4.1 Services
The application layer is the layer that provides communications functions for
a network application to serve an end user or another application program.
If an application layer entity is providing a service direct to an end user, then
the software that provides the interface between the end user and the
networked application is described as a user agent.1
Network applications are often implemented as client server systems.
Under the client server model, the client only runs when it is required and
initiates a request to the server and the server replies with a response.2 The
server will typically handle requests from many clients and will run
continuously. Both the user agent (client) and the server run application
processes that work together via a network to deliver the application service
to the end user.
The precise service offered by the application layer will vary from application
to application, but will often involve identification of the communicating
partners and the agreement of the responsibility for error recovery, security
aspects and data encoding. The application layer is also responsible for
negotiating and meeting certain quality of service requirements for reliable
data transfer, throughput or for delays. Some applications are loss tolerant
while others are loss sensitive. Some applications are bandwidth sensitive
1
A web browser such
as Internet Explorer or
Netscape Navigator is
an example of a user
agent. It provides the
interface between an
end user and the world
wide web application.
Email clients are also
examples of user
agents.
2
Both web browsers
and email clients
conform to the client
server model.
29
while others are elastic in their bandwidth requirements. Similarly some
applications are delay tolerant while others are delay sensitive. Tanenbaum
provides a useful table3 that describes the quality of service requirements for
different network applications.
4.2 Interfaces
For user agents, the interface to the application layer is today likely to be a
Graphical User Interface (GUI), such as that provided by the Windows
operating system. It could also be a command line interface, as provided by
DOS or Unix.
Application layers may also provide services to other applications by
means of an Application Programming Interface (API), which will
provide a set of library functions that can be called from application
programs.
4.3 Functions
The main function of the application layer is to organise the necessary
resources to allow an application process on one system to communicate
with an application process on another system via a network. The application
layer may also synchronise the application processes at both ends so that
they can communicate successfully.
The application layer will often provide reliable communications to the
application processes, especially when the application layer makes use of an
unreliable transport service. The application layer is the layer of last resort
which must correct all the problems that have not been dealt with by the
lower layers. The application layer must therefore, unless it is using a reliable
transport service, be able to detect the loss, corruption and duplication of
messages and be able to recover from these problems. It must also be able to
control the flow of data if the receiver or the network cannot handle the rate
of data being transmitted.
The fundamental decision that designers of applications must take is which
transport service to use. It is the transport service that supports the differing
levels of service that might be required. For internet applications, there is a
choice between a reliable service using the Transmission Control Protocol
(TCP) or an unreliable or best efforts service using the User Datagram
Protocol (UDP). The reliable service uses a complex connection oriented
transport protocol, and the unreliable service uses a simple connectionless
transport service. The choice of what sort of transport service to use will
have a huge effect on what functions are required in the application layer. It
may seem strange at first to think that some application developers would
prefer to use an unreliable transport service, but there are several reasons
why this might be the best choice.
• Real time applications are more tolerant to packet loss than they are to
delay. Losses of occasional packets will not make much difference to
audio or video transmissions, as losses are relatively infrequent and the
applications can interpolate missing data, so that losses can be hidden
from users. Delays, however, and particularly variable delays, do cause
problems that can be observed by users. They can result in a jerky effect
which is quite disconcerting. Using TCP, as we will discover later, does
give rise to extremely variable delays, while delays with UDP are less
severe and are more consistent.
30
3
Tanenbaum, Figure 5–30,
page 397.
• The client server model is well suited to using a connectionless transport
service. If clients make occasional request to servers, then using a reliable
connection-oriented service, such as that provided by TCP, can be very
inefficient. It will be necessary to set up a connection and close it down
afterwards. This will require a minimum of five packets and the server
will have to hold state information in its memory about all the transport
connections that are currently active. Communications operate much
more efficiently, as do applications, if a connectionless transport service is
used. There will only be a need for two packets to be exchanged and the
server can minimise memory usage as it does not need to hold any state
information about connections. If packets are lost or corrupted the
application client simply retransmits its request.
• If applications are very security conscious, they will not trust the
transport service or anything else that was developed or is managed by
other parties. The designers of such applications will want to detect and
recover from errors within the application itself. In this case, it would be
pointless to replicate this functionality in the transport layer and it would
be much more efficient to use an unreliable transport protocol.
• The implementation of a reliable transport service requires a large
amount of complex code. Simple devices without much memory or
processing capability are unlikely to be able to support a reliable transport
service. The memory and processing requirement overhead for a reliable
transport service may force application designers to choose a
connectionless transport service.
Some applications that make use of connection-oriented transport services
may also allow the grade of service to be requested and negotiated when the
connection is established. The parameters that are of interest relate to
reliability, throughput and delay.
4.3.1 Encapsulation
Data from end users or other applications is encapsulated in an application
layer PDU by prefixing the data with an application layer header specific to
the application protocol.4
4.3.2 Addressing
Addressing is often thought of as a function of lower layer protocols, but
many application protocols do require their own addressing function as well
as having to pass down addresses to be used by lower layer protocols.5
4.3.3 Ordered delivery
Ordered delivery is a function of a reliable transport or network service, but
where an application is using an unreliable transport service, PDUs can be
received in the wrong order as they can take different routes through the
network. Where this happens, the application layer protocol must contain a
sequence number field so that the application layer can determine if PDUs
arrive out of order. The sequence numbers can also be used to request that
lost PDUs are retransmitted, and to reorder them if necessary so that they
can be handed in order to the appropriate application process or user agent.
To do this the application layer needs to buffer the PDUs received, and
therefore a certain amount of memory must be allocated for storing
incoming PDUs while earlier PDUs are awaited.
4
For example, an email
address message body
is encapsulated with
Simple Mail Transfer
Protocol headers
containing fields such
as the FROM field and
the SUBJECT field.
5
For example, when
requesting a web page
using Hyper-Text
Transfer Protocol
(HTTP), the ‘address’ of
the page will be
indicated by the
Uniform Resource
Locator (URL) field. The
URL will indicate the
application layer
protocol to be used
(HTTP), the name of the
host from which the
web page is to be
fetched (network layer
addressing), the port
number to be used
(transport layer
addressing) as well as
the file to be
transferred (application
layer addressing).
31
4.3.4 Flow control
Applications which do not make use of a reliable transport service will also
require an end-to-end flow control mechanism so that the receiver can
regulate the flow of data from the transmitter. To do this the application
protocol header will require a sequence number and an acknowledgement
field so that the receiver can acknowledge each PDU transmitted. It can then
slow the rate of transmission by not acknowledging PDUs until it is ready to
receive some more.
4.3.5 Error control
Where an application is using an unreliable transport service (or where the
application does not trust a reliable transport service), the application must
perform its own error detection and recovery. This will require a redundant
error checking field in the application protocol header as well as sequence
number and acknowledgement fields, so that the application layer can check
that all the PDUs have been received and so that retransmission can be
requested and PDUs re-ordered if necessary.
4.3.6 Connection control
The connection control function is required in all application protocols
that make use of a connection-oriented transport service. The application
layer must be able to establish connections prior to transmitting data and
to close them when there is no more data to be transmitted. The
application layer must identify and determine the availability of the
application processes which wish to communicate and establish their
authority to do so. It must also determine the mode of communication
(simplex, half duplex or full duplex). A facility is also required so that
connections can be reset or reinitialised to a known state should serious
problems be encountered.
4.3.7 Security
Security is often an important function of the application layer, as many
applications will assume that all networks are insecure, and application
layers sometimes prefer to implement security at this level, rather than
make use of security functions within the transport and network layers.
Schemes are required to authenticate the parties involved in the
communication and to prevent any other parties from being able to read
or alter the data being transmitted.
4.3.8 Data encoding
An important function of the application layer is to determine how data is to
be encoded for transmission. This may involve the choice of character codes,6
the use of tags to define how data is displayed7 or to define data syntax or
semantics.8
Abstract Syntax Notation 1 (ASN.1) has become an important standard for
specifying PDU formats at all layers. It is an ISO standard, but it has also been
used extensively by the IETF in specifying new Internet based protocols. It
can be thought of as a type definition language where data is defined as
belonging to either primitive types such as Boolean, Integer or Bitstring, or
more complex user-defined types. It is similar to data type declarations in
programming languages. ASN.1 is not only used for specifying protocols
32
6
For example, American
Standard Code for Information
Interchange (ASCII), Unicode and
Extended Binary Coded Decimal
Interchange Code (EBCDIC).
7
For example, Hyper-Text Markup Language (HTML).
8
For example, Extensible Markup Language (XML) and Abstract
Syntax Notation 1 (ASN.1).
(abstract syntax), it is also used to code the data that is being transmitted in
the PDUs (transfer syntax). It does this by using a set of Basic Encoding Rules
to code the data types and the data.
4.3.9 Data encryption
Data is often encrypted between application layer entities to ensure that it
cannot be viewed or altered by third parties as it is transmitted across
networks.
4.3.10 Data compression
Data compression is often required because bandwidth in wide area
networks is a scarce (and hence expensive) resource, and some types of data
(such as voice and video) require a large amount of bandwidth, although
they use that bandwidth quite inefficiently. Typically voice and video signals
(and to some extent also text) contain a large amount of redundant
information and can be coded much more efficiently using data compression
algorithms. These algorithms can either be lossy,9 where information cannot
faithfully be reproduced at the receiver, or they can be lossless,10 where the
quality of the information after decompression at the receiver is just as good
as it was before it was compressed at the transmitter.
There are many different complex data compression algorithms used for
coding data prior to transmission to conserve network capacity. All of them
apart from Huffman Coding are beyond the scope of this syllabus. Huffman
Codes use variable length codes for different symbols depending on how
frequently they are used. The ASCII character set uses 8-bit codes (including
a parity bit) to define all the characters of the alphabet and other characters.
But some characters occur much more frequently than others. Huffman
Codes allow more frequently used characters to be represented by fewer bits,
and less frequently used characters to be represented by more bits. By doing
this a significant reduction can be achieved in transmitting a large amount of
text. But, if characters are represented by variable length codes there must be
a clear method for determining the start and end of the code for each
character. Huffman Codes do this in a clever way by coding each character as
the path from the root to a leaf of a binary tree called a Huffman Tree. Thus
messages can be encoded in an unambiguous way, so that the receiver can
always decode the message and knows that when it reaches a leaf node, a
character has been received.
9
E.g. Joint
Photographic Expert
Group (JPEG)
compression of still
images and Motion
Picture Expert Group
(MPEG) compression of
video.
10
For example,
Huffman coding of text.
Figure 4.1: Example of a Huffman Tree to encode the word MESSAGE
0
0
1
1
0
1
A
1
0
G
M
1
E
S
33
The Huffman Tree in Figure 4.1 can be used to encode letter M by navigating
down from the root to the leaf node M and writing down a 0 whenever a left
branch is taken and a 1 whenever a right branch is taken. M is therefore
encoded as 100. Similarly E is encoded as 11, S as 101, A as 00 and G as 011.
The word MESSAGE is therefore encoded as 100111011010001111.
The bit string above can be unambiguously decoded by starting at the root
and taking a left branch whenever there is a 0 and a right branch whenever
there is a 1 until a leaf is reached which signifies the coded letter. The next
letter is decoded by the same method starting again at the root.
A Huffman code, although designed from a tree, can also be defined in a
table. The Huffman code defined in the tree in Figure 4.1 could also be
defined by means of Table 4.1.
Table 4.1: Example of a Huffman Code table to encode the word MESSAGE
A
00
E
11
G
011
M
100
S
101
From a Huffman Code table, a tree can be drawn by placing each letter as a
leaf of the tree according to the navigation rule for the code (left branch for
a 0 and right branch for a 1).
For codes involving many letters, it is easier to code from the table and to use
the tree for decoding.
4.4 Virtual terminal protocols (Telnet,11 SSH and VT)
4.4.1 Telnet
Telnet was one of the original Internet application protocols. It is an example
of a virtual terminal protocol that allows users on a character-mode terminal
(or more commonly these days on a PC running a terminal emulator) to log
into and execute commands on a remote host using a command line
interface. It uses a reliable TCP connection. Telnet just relays any characters
typed by a user to the remote host, and allows commands to be entered at
the remote host as if they came from a local terminal. It then relays any
characters sent in response back to the user. These characters will include
any user name and password requests, but it should be noted that passwords
will be transmitted as plaintext and the protocol therefore is not secure.
Telnet can support many different terminal types and translate between
different character codes, if necessary. It does this by translating to a
standard format, known as Network Virtual Terminal (NVT) for
transmission across the network.
Telnet command codes can be embedded in the data stream. To achieve this,
a special escape character (FF in Hexadecimal) is required so that the
receiver knows to interpret the next character(s) as a command code.
Embedding control information like this within data is known as in-band
signalling. Also, the user has sometimes to enter a command to Telnet to
request the Telnet client to close a connection, for instance, and an escape
character (usually CTL-SHIFT-6) is needed so that the next line is interpreted
34
11
RFCs 854–861.
as a Telnet command to the client and not as data to be sent to the server.
This illustrates a problem with in-band signalling, that data transmission will
not be completely transparent, as some combinations of characters cannot be
sent as they may be interpreted as commands.
Telnet messages do not have application layer headers. They consist of the
characters being typed by the user or sent by the server together with
occasional command codes generated by the Telnet client or server.
A further potential problem with Telnet that makes it very inefficient over
WANs is that it was designed for use over asynchronous modem links. The
standard method of remotely accessing a host at the time was for a character
to be sent to the host and the host to echo it back to the terminal before it
was displayed. This method of operation is known as echoplexing. It
required a full-duplex link, but had the advantage of showing users when
noise on the line was corrupting data. It is not well suited to modern packetswitched networks because there is a delay between typing a character and
seeing it appear on the screen, and each character will be transmitted to and
from the host in a single packet, incurring large protocol overheads. Also,
with modern WAN links, the probability of characters being corrupted is
quite low. Because of this, it is advisable to turn off the echo function at the
remote server and enable a local echo at the client.
Telnet is not commonly used today because of security concerns, but it is still
often used by network managers to remotely manage and configure routers.
Network managers can make Telnet much more secure by configuring the
routers to only accept Telnet sessions from known IP addresses,
corresponding to the network management workstations.
4.4.2 SSH
Another way to improve security for remote login type applications is to use
Secure Shell (SSH). The functionality of SSH is very similar to Telnet, but
user names and passwords are encrypted for transmission from SSH clients
to SSH servers.
4.4.3 VT
Virtual Terminal (VT) is the ISO equivalent protocol that offers similar
functionality to Telnet, but like many other ISO protocols, it has found it
difficult to compete with the protocols designed for use on the Internet.
Activity 4.1
Attempt to access a host at your university/institution from a PC or another Unix host
using Telnet by entering telnet hostname at a DOS/Unix command prompt. It may be
that Telnet access to hosts will be barred for security reasons. If this is the case,
investigate whether SSH is supported and if so, attempt to establish an SSH connection
to a remote host.
Then do the same from a web browser by entering telnet://hostname in the address
window of the browser.
4.5 Web protocols (HTTP12 and HTML)
12
RFCs 1945, 2068.
The world wide web, Hypertext Mark-up Language (HTML) and the
Hyper-Text Transfer Protocol (HTTP) were all invented by Tim Berners-Lee
at CERN in 1989. The world wide web is now the most important and widely
used Internet application.
35
HTTP and HTML were studied in a Level 1 unit. You are strongly advised to
revise this material, as the remainder of this section on web protocols will
assume this knowledge and build upon it.
HTTP assumes a client server model for communications. The web browser
acts as the client and requests web pages from the web server. The page is
referenced by a Uniform Resource Locator (URL) which can be thought of
as an application layer address. It defines the protocol to be used, the
location of the server and the file to be transferred as well as the transport
layer address to be used.
A URL has the following structure:
protocol://hostname/filename:port number13
The protocol is usually HTTP (in lower case), but browsers do support other
protocols, such as FTP and Telnet. The URL can also be used to access a local
file by specifying “file://” as the protocol followed by the full directory path
for the file. The protocol field of the URL is always terminated by “://”
The hostname is the domain name of the host and is a series of hierarchical
domain names in ascending order separated by “.”. By convention, the
hostname of most web servers usually start with “www.” as the lowest level
in the hierarchy of domains. The filename includes a full directory path with
directories being delimited by “/”s. Finally, the port number (which is
optional) is preceded by “:” and is used to indicate the transport layer
address that is to be used. HTTP employs TCP for its transport service and, by
default, will use port 80 which is the registered port number for HTTP.
The HTTP protocol is actually quite simple, as there are only a small number
of basic operations, called methods. The most common is the GET Request
and Response that fetches a web page from a server. At its most basic the
request header could just contain a field to indicate the method, the URL of
the required web page and the version of HTTP to be used. The response will
normally be the requested file, but the response header also contains a status
code. Some of these status codes will be displayed by the browser in the
event of a failure.14 In HTTP Version 1.0, a TCP connection was opened for
each object to be transferred and closed afterwards. These are known as
non-persistent connections. In HTTP Version 1.1 and all subsequent
versions, persistent connections were supported and many objects could be
transferred over the same connection, thus improving response times.
Response times were further improved in HTTP 1.1 by allowing the client to
issue new requests before responses had been received to earlier requests.
This facility is known as pipelining. The default mode of HTTP 1.1 is to use
persistent connections with pipelining.
HTTP has a simple security mechanism that developers can implement to help
prevent unauthorised access to web pages. A web page can be set up so that
authorisation is required. In this case the web server prompts the client for a
user name and password. The server requests a user name and password with
a 405 Authorization Required Response and the browser prompts the user for
this information. Once the browser has obtained a user name and password it
resends the request but this time includes the user name and password in the
request header. The server will then check this and if satisfied, will download
the page. The browser will cache the user name and password and will
automatically include them in any further requests to the server during that
browser session.
A secure version of HTTP called Secure HTTP (HTTPS) has been
implemented which is identical to HTTP but runs over a secure transport
connection provided by the Secure Sockets Layer (SSL) which is
36
13
For example,
http://doc.gold.ac.uk/~mas01pt/c
is 222:80 indicates that the file
~mas01pt/cis222/index.html
(index.html is the default name of
the file to be retrieved) is to be
requested using HTTP from the
Goldsmith’s Department of
Computing Web Server on the UK
academic network
(doc.gold.ac.uk) using port 80.
14
For example, 200 OK, 301
moved permanently, 400 Bad
Request, 401 Authorization
Required, 404 Not Found and
505 HTTP Version Not Supported.
implemented between the application layer and the transport layer and
certifies the identity of the web server. You can tell when you are using
HTTPS because the URL will be commence with https:// in the address
window of the browser and a padlock icon will appear at the bottom of the
browser window. Details of the certificate can be obtained by double clicking
the padlock icon.
Activity 4.2
Attempt to set up a Telnet connection to a web server at your university/institution or
another other web server on the Internet using the HTTP port (80). The DOS/UNIX
command line syntax for this will be telnet hostname 80.
For example, type telnet websiteaddress 80 to establish an HTTP session with a host
web server that you know. Then once the connection is established, type GET
pagename HTTP/1.0.
The pagename can be index.html for the default home page of the site or it could be a
structured hierarchical name. This will fetch the HTML code for the appropriate page
from the web server.
4.6 Mail protocols (SMTP15, MIME16, POP317, IMAP18 and
MOTIS)
Electronic Mail was also one of the earliest Internet application protocols to
be designed. It is a store and forward text messaging protocol supporting
mail clients (user agents that send and receive messages) and mail servers
that relay messages to each other and to and from mail clients.
15
RFC 821.
16
RFCs 2045–2049.
17
RFC 1939.
18
RFC 1730.
Email clients have five basic functions:
• composition, which allows users to create messages
• transfer, which allows users to transfer messages to and from the mail
server
• reporting, which allows the mail server to indicate such things as a
message not having been delivered
• displaying, which allows the mail client to display the headers of
messages and their contents
• disposition, which allows the user to delete messages or store them in
folders.
Email clients use two different protocols. One for sending messages to the
mail server and one for retrieving messages from the mail server.
4.6.1 Simple Mail Transfer Protocol (SMTP)
Simple Mail Transfer Protocol (SMTP) is used to transfer messages from a
mail client (user agent) to a mail server (mail transfer agent) and is also used
to transfer messages between mail servers. It is a very simple text based
protocol. Messages comprise a set of headers and a body. There are two
envelope headers which start with MAIL FROM:, used to identify the
message originator and RCPT TO:, used to indicate the recipient(s) of the
message. Unlike many more modern protocols, each envelope header is
transmitted and acknowledged separately, rather than encapsulating the
body. The message itself is prefixed by the keyword DATA followed by the
text to be transmitted, but this also has its own headers such as FROM:, TO:,
SUBJECT: and DATE: from which the envelope addresses are obtained.
These headers are followed by a blank line and then the actual text of the
message. The body is terminated by a new line with just a full stop on it and
37
then another new line. All bodies and headers are coded in 7-bit ASCII text.
SMTP is not a real-time protocol. Messages are stored at clients and servers
and forwarded at regular intervals using reliable TCP connections. SMTP
PDUs do not really have application layer headers as normally understood.
They consist of keywords followed by some data.
SMTP does not offer any guarantees about delivery of messages, although it
is quite robust and considered to be reliable.
4.6.2 Multipurpose Internet Mail Extensions (MIME)
Because SMTP was designed only to carry ASCII 7-bit characters, it cannot
on its own be used to transfer 8-bit binary data which would be required if an
executable file or a formatted text file (such as a Word document) were to be
transmitted. When the designers of SMTP realised that users wanted to the
ability to send data other than ASCII text, they were faced with two possible
solutions. Either they could change the SMTP protocol and update all the
clients and servers at the same time, or they could just create a new protocol
to allow 8-bit data to be encoded as 7-bit ASCII characters. They chose the
latter option, as it only involved upgrading the clients, and they designed a
new protocol called Multipurpose Internet Mail Extensions (MIME) which
allowed 8-bit data files to be attached to SMTP messages and be transmitted
as 7-bit ASCII characters.
MIME defines a number of standard data types and sub-types.19 These MIME
data types have become a standard in many Internet and other applications.
SMTP assumes that mail servers operate continuously and are always
available. If they are not available, the messages will be stored and forwarded
when the mail server becomes available. Clients, however, are frequently not
available, as users do not keep their mail clients running all day and often do
not have a permanent connection to the Internet. For this reason SMTP is not
well suited for delivering messages to clients. Instead, other protocols were
designed to allow clients to connect to servers and request that messages are
downloaded. Because these protocols were designed to work over dial-up
networks, they also require security mechanisms to ensure that mail is being
downloaded by valid users.
4.6.3 Post Office Protocol 3 (POP3)
One protocol that does this is Post Office Protocol 3 (POP3). The protocol
has three phases. It has an authentication phase where the user is
authenticated by a user name and password; a transaction phase where
messages are downloaded from the mail server; and an update phase where
the messages on the server will be deleted (if required) after they have been
successfully downloaded to the client.
4.6.4 Internet Mail Access Protocol (IMAP)
An alternative protocol for retrieving messages from mail servers is Internet
Mail Access Protocol (IMAP) which offers very similar functions to POP3,
but also allows users to view message headers and select which messages to
download. This is very useful for a dial-up connection where bandwidth is
expensive and should not be wasted by downloading spam or other
unwanted messages. IMAP also allows messages to be stored in folders on the
server, which is particularly important if the user often accesses his email
from different machines.
38
19
For example, text/plain,
text/html, image/jpeg,
application/msword,
application/pdf, audio/basic and
video/mpeg.
4.6.5 Hyper-Text Transfer Protocol (HTTP)
Another, and increasingly popular method for sending and receiving email
messages is to use the world wide web to access mail services such as
Hotmail, making use of the web’s Hyper-Text Transfer Protocol (HTTP)
rather than using mail protocols. As with IMAP, messages can be organised in
folders on the server. Web-based email has the advantage of allowing users
to access their email from any machine that supports a web browser, such as
a PC in an Internet Café.
4.6.6 Message-Oriented Text Interchange Standard (MOTIS)
Message-Oriented Text Interchange Standard (MOTIS) is an ISO
messaging standard. It is based on the ITU-T X.400 Message Handling
Service (MHS) standard. This standard, unlike SMTP, is very complex and
sophisticated and does many things that SMTP does not do. It has not been
very successful and virtually all email today still uses SMTP. MOTIS/X.400 is
perhaps too complex and users prefer the simplicity of SMTP, particularly
with regard to email addresses. X.400 addresses consist of a set of keywords
and values which are much harder to remember and more cumbersome to
use than SMTP addresses.20
Activity 4.3
Attempt to connect to the mail server at your university/institution or your ISP’s mail
server using the SMTP port (25). The DOS/UNIX command line syntax for this will be
telnet hostname 25. You should be able to find the domain name for the mail server
by looking at the configuration of your mail software. The port number is a transport
layer address which will be discussed in the next chapter.
20
As an X.400 address,
p.tarr@gold.ac.uk
would appear as, I=P;
S=TARR; O=GOLD;
PRMD=UK.AC; ADMD=
; C=GB.
You will now be able to enter the following SMTP commands in bold and you should
receive the SMTP responses in ordinary type.
MAIL FROM: <your email address enclosed in angle brackets>
250 <your email address enclosed in angle brackets> is syntactically correct
RCPT TO: <your email address enclosed in angle brackets>
250 <your email address enclosed in angle brackets> is syntactically correct
DATA
354 Enter message, ending with “.” on a line by itself
From: <your email address>
To: <your email address>
Subject: Test
This is a test.
.
250 OK id=<message id provided by mail server>
QUIT
If you follow the above procedure, your mail server should deliver a message with the
subject “Test” to your mailbox. Note that the DATA field which also contains some mail
headers is terminated by a line that contains just a full stop. Using Telnet to mimic SMTP
works because Telnet allows TCP connections to be set up to any port number (not just
on port 23 that Telnet uses by default) and because it allows ASCII characters to be
transmitted transparently without any protocol headers. The mail server is also expecting
to receive ASCII characters but only on TCP connections using port 25.
39
4.7 File transfer and access protocols (FTP21, TFTP22, NFS23
and FTAM)
21
RFC 959.
22
RFC 1350.
23
RFC 3530.
4.7.1 File Transfer Protocol (FTP)
File Transfer Protocol (FTP) was also one of the original Internet
application protocols. It allows three different types of file (unstructured,
structured and random) to be transferred over a network between one host
and another using a set of simple commands. File transfers are now quite
often carried out using HTTP rather than FTP, but FTP is still used not least
when web pages are published to a web server. Most web publishing
software uses FTP, although the details of it are hidden from the users and
few web designers will be aware that they regularly use FTP.
The original FTP implementations were command line interfaces for use on
Unix hosts, and allowed users to view directories on remote hosts and the
files they contained, change directories when necessary using Unix
commands and then download or upload files between two hosts.24 FTP
converts these user commands to standard FTP commands (three or four
letter codes) which are sent via the control connection and elicit responses
containing a three digit status code followed by some text from the server.
Modern FTP implementation uses a graphical user interface where the local
and remote directories can be viewed together and files to be transferred can
be highlighted and then transferred by clicking an arrow button that
indicates the direction of the transfer.
The protocol actually makes use of two TCP connections: one for control
commands and responses, and the other for actual file transfer. The control
connection is left open for a whole FTP session, while the data connection is
established to transfer a file and closed as soon as the file has been successfully
received. This method of control using one connection for control and another
for data is called out-of-band signalling as opposed to other protocols such as
Telnet and HTTP which use in-band signalling. Because all control
information in FTP is passed via the control connection, the data connection
does not require any application layer headers and is a simple TCP connection.
FTP can convert between different character codes. It converts data to the
same Network Virtual Terminal (NVT) codes as Telnet for transmission.
Unlike HTTP, FTP cannot be a stateless protocol. The FTP server has to
remember which connections belong to which FTP users, and also which
current working directories each FTP user is using.
The connection used for commands actually uses the Telnet protocol to
transfer commands and responses to the remote host. For security purposes,
hosts normally require a user name and password to be entered for all Telnet
sessions. But this would hinder general public FTP access to a site, as users
would have to pre-register. A convention that developed for public FTP access
was for public FTP servers to accept the user name ‘anonymous’, but not to
perform a password check. Instead, anonymous users are expected to enter
their email address as the password, so that the host can, if it wants, collect
some details on the users of its public FTP service.
4.7.2 Trivial File Transfer Protocol
There are some situations where a complex file transfer protocol is
inappropriate. Some network devices, such as low-cost routers, do not have
sufficient memory and processing capability to justify the implementation of
40
24
Typical FTP user commands
include open to establish an FTP
command connection to a host,
close to close an FTP command
connection, ls to list the contents
of a remote directory, c d to
change a remote directory, get to
open a data connection and
transfer a file from the remote
host, put to open a data
connection and transfer a file to
the remote host, and quit to
close FTP
such a protocol, if it has to sit on top of a complex connection-oriented
transport protocol. Also, between two hosts on the same LAN the probability
of errors is quite remote and it may be preferable to use a lighter weight file
transfer protocol. A disc-less work station, which has to download all its
software from a server over a LAN, is a good example of a situation where
only a light-weight file transfer protocol is desirable. Trivial File Transfer
Protocol (TFTP) is such a protocol. As its name implies, it is extremely
simple. It uses UDP as its transport service and provides quite a thin
application layer. Each application layer message has to be explicitly
acknowledged before another message can be sent. Each TFTP message
carrying data contains an application header that includes a sequential block
number. The receiver will then acknowledge the receipt of the message with
an acknowledgement message that contains the block number. The
transmitter will re-transmit a message if an acknowledgement is not received
before a timeout expires. Unlike FTP, TFTP only supports file transfer. It does
not support any interaction to locate files in directories. The files and the
direction of transfer are specified in the command line that is used to call
TFTP. TFTP also has no facility for authenticating users. For security
purposes, network managers should therefore only allow TFTP traffic to and
from known IP addresses. TFTP is often used by network managers to
download or upload router configurations and software.
4.7.3 Network File System (NFS)
Sometimes it will be more efficient to access files remotely rather than to
transfer them in their entirety. A popular means of doing this is to use the
Network File System (NFS), originally developed by Sun Microsystems for
the Unix environment, but which has since been ported to most other
commonly used environments. NFS allows physically remote directories to be
mounted on local systems, so that the directories and their files appear to be
local to the users. All the standard operations that are carried out by the
Operating Systems on local directories and files are supported transparently
on the remote directories and files. NFS is implemented using an application
mechanism called Remote Procedure Calls (RPCs), also developed by Sun,
where software that normally calls procedures on the local system can call
equivalent procedures on the remote system. RPCs are implemented using a
very simple protocol that packs the name of the procedure and any
parameters required into a message using a coding system called External
Data Representation (XDR). This is sent as a request to the remote system
which unpacks it and calls the procedure. It then packs the return value and
any other output parameters into a response message to be sent back to the
calling system, which then returns these to the original calling process. NFS
can use either TCP or UDP for its transport service but, because it is a simple
client server application, it is best implemented on top of UDP. NFS is not
very secure, but authentication services have been developed that offer
improvements in this area.
File Transfer Access and Management (FTAM) is the equivalent ISO
protocol to FTP. It is not in common use today.
Activity 4.4
Use the file transfer protocol on a PC or on a Unix system at your university/institution to
transfer files to and from an FTP server at your university by entering ftp hostname at a
DOS/Unix command prompt, or by using a windows-based version of FTP. The domain
name for FTP servers begins, by convention, with “ftp.” just as the domain name for a
web server, by convention, often begins with “www.”. Use the same software to perform
an anonymous file transfer from a public FTP server by downloading some RFCs from
41
ftp://ftp.rfc-editor.org. Then transfer files from the same FTP servers using a web browser
by entering ftp://hostname in the address window of the browser.
4.8 Directory protocols (DNS25, X.500 and LDAP26)
25
RFC1034, 1035.
26
RFC 3377.
4.8.1 Domain Name System (DNS)
Access to directories is required by a number of network functions, as well as
by other applications. A directory service called the Domain Name System
(DNS) was developed for the Internet to allow applications to use host
names, and then for these host names to be translated (or resolved) into
network layer addresses. In the early days of the Internet, the mapping
between flat host names and network layer addresses was done via a text file
that was centrally managed and then distributed to all the hosts on the
Internet. This method soon became too unwieldy as the Internet grew in size,
and an alternative method using a hierarchical, fully distributed system was
devised called the Domain Name System.
The DNS required that all host names were hierarchically structured. It can
be viewed as a tree with a single root. The top layer of the hierarchy is the
last field in the hierarchical name. It can be .com, .org, .edu, .net, etc. or it
can be a country code such as .in, .hk, .si, .uk, etc. These domains are
allocated by the Internet Corporation for Assigned Names and Numbers
(ICANN) which is responsible for managing the thirteen top DNS root
servers. For each of the top level domains there are at least two DNS servers.
Below this level there are second level domains which manage the
authoritative DNS servers for the domain, and have delegated authority to
manage all the domains levels below them. They might well delegate further
levels to other organisations. This delegation can continue to the third or
sometimes the fourth level.27
When an application needs to resolve a hostname it calls a name resolving
function to issue a DNS request to its local DNS server. The network address
of the local DNS server is configured in the client computer. The local DNS
Server will hold data on all domains for which it is authoritative as well as
holding a cache of previously resolved names. If the local server is able to
resolve the name, it will return the network layer address in a DNS response.
If the server is not able to resolve the name, it will make a request to one of
the root servers. The root server will return the correct network layer address
if the name is held in its own data or cache, otherwise it will return the
network layer address of an authoritative server that can help resolve the
name. This process could repeat itself a number of times until either the
name is resolved or it can be firmly established that the name cannot be
resolved. This process can be either recursive, where the request is passed on
from server to server and the responses passed back along the chain of
servers, or iterative, where a server makes a number of requests in turn to
different servers. DNS can operate over either UDP or TCP transport
protocols. Because it is a very simple client server application, virtually all
implementations of DNS use UDP for efficiency reasons.
DNS is also used by Internet mail servers to resolve the domain name part of
email addresses to the network layer address of the destination mail server.
4.8.2 X.500
The ITU-T and ISO have also developed a comprehensive and complex
directory service and protocols. Their directory system is often known by its
ITU designation X.500. It operates between Directory User Agents (DUAs)
42
27
For instance, the host name
doc.gold.ac.uk has the UK as its
top level domain; the Academic
Community is second level
domain; Goldsmiths as its third
level domain; and the
Department of Computing as its
fourth level domain. ICANN has
delegated the authority for
allocating all .uk domain names
to the UK registration authority
(Nominet) which has delegated
authority for allocating .ac.uk and
.gov.uk domain names to the UK
Education and Research
Networking Association
(UKERNA) who have allocated
the gold.ac.uk name to
Goldsmiths College who have
then allocated the doc.gold.ac.uk
domain to the Department of
Computing.
and Directory Service Agents (DSAs). The protocols used are the Directory
Access Protocol (DAP) which is used between DUAs and DSAs and
Directory Service Protocol (DSP) used between DSAs. X.500 like X.400 is
based on an object-oriented design and it also incorporates ASN.1 coding.
Although X.500 was originally designed as the directory service for resolving
names into X.400 mail addresses (a much needed service given the
unfriendliness of X.400 mail addresses), it was designed to be much more
generic than DNS and can support directories for any type of object stored in
a Directory Information Base (DIB). DAP and DSP make use of a number of
different ISO protocols in each layer of the ISO protocol stack.
4.8.3 Light-weight Directory Access Protocol (LDAP)
The X.500 directory service, like most other ISO protocol developments,
could not compete with its Internet equivalent. However, the design and
many of the ideas behind X.500 are good and there is a need for a welldesigned directory service that companies can use to look up employees’
phone numbers, email addresses and office locations. With this in mind, the
University of Michigan adapted X.500 to create a less complex protocol called
Light-weight Directory Access Protocol (LDAP), which, despite its name, is
still not particularly light-weight. LDAP is used in Microsoft’s Active Directory
and Novell’s Netware Directory Service products.
Activity 4.5
Try to find a copy of a program called nslookup (Name Server Look-up). It can be found
on most Unix systems, but is not usually included with Windows.
Use nslookup to resolve some hostnames. In Unix this can be done on a single command
line by entering nslookup hostname at a Unix command prompt. If the name can be
resolved, the name and network address of the name server will be displayed, possibly
followed by a line that says “Non-authoritative answer:”. This means that the name was
found in cache. Finally, the name and network address of the host are displayed. If the
hostname you look up is obscure enough, you might get an authoritative response that
does not include the “Non-authoritative answer:” line.
It is more interesting to run nslookup in debug mode. This can be done by entering the
nslookup command without any hostname and then typing set debug before entering a
host name on a new line. You will now see more information on the authoritative servers
for the host’s domain.
If you want to see a verbose account of everything nslookup is doing, there is a second
debug mode you can enter by typing set d2.
There are many other options supported by nslookup. You can read about them by typing
man nslookup at the Unix command prompt.
You can return from nslookup to the Unix command prompt by typing exit.
An alternative Unix tool for name resolution and DNS debugging is dig (Domain
Information Groper). Read about dig and its options by typing man dig and then try to
use it to resolve some host names.
4.9 Network Management Protocols (SNMP28 and CMIP)
28
RFC 3416.
Network management protocols is another area where there was a battle
between Internet and ISO standards which was conclusively won by the
Internet standards. Yet again the Internet developers chose to launch a
simple but effective network management protocol which could be
implemented quickly, while the ISO developers created a technically complex
protocol that had many more functions and would take a long time to
implement and debug.
43
The Internet Network Management standard is known as Simple Network
Management Protocol (SNMP) and the ISO standard is known as Common
Management Information Protocol (CMIP).
SNMP runs on top of UDP, as it only has to support a small number of
request/response interactions to network devices that often have limited
memory and processing capability. The protocol does however use the ISO’s
ASN.1 transfer syntax for encoding data. It operates between a network
management station and a management agent within the network device
that collects management information in a Management Information Base
(MIB). The network management station can request the value of any object
in the MIB and can also set the value of an object in the MIB. The network
device can also issue a trap to notify the network management station that a
problem has occurred.
SNMP version 2 is not very secure as passwords (known as community
strings) are transmitted as plaintext. SNMP version 3 addresses these
security problems.
CMIP is a complex protocol that makes use of a number of different ISO
protocols in each layer and, as a result, is not particularly well suited to
managing simple devices which would require considerable memory and
processing power to support all the protocols. It does have good security
features.
Network management protocols and network management in general will be
studied in more depth in the second half of this course.
i. Post Office Protocol 3 allows users to view email headers without
downloading the messages’ bodies and is the recommended protocol
for use with low speed dial-up Internet access.
ii. Telnet runs efficiently on top of UDP when used with asynchronous
character-mode terminals.
iii. FTP requires that two transport connections are set up in order to
transfer a file.
iv. Simple Network Management Protocol, which was standardised by the
IETF, encodes data in ASN.1 which was standardised by ISO.
b. Describe the format of a Uniform Resource Locator.
c. Using the Huffman Code defined in Table 4.1, code the word GAMES. Draw the
Huffman Tree and describe how the Huffman code for GAMES can be
unambiguously decoded from the Huffman tree.
d. What does the acronym MIME stand for? Why is it necessary for all mail agents
to implement it?
e. Describe how DNS resolves a host name into an IP address.
44
Learning outcomes
At the end of this chapter and the relevant reading, you should be able to:
• describe the services and interfaces offered by the application layer
• describe the functions of the application layer
• identify the reasons why application developers might choose to use an
unreliable transport service and how this choice will affect the design of the
application layer protocol
• outline how ASN.1 is used to encode data
• encode and decode compressed messages using Huffman codes and trees
• describe how Mail protocols transmit and receive emails
• identify the main differences between POP3 and IMAP in retrieving mail from
servers
• describe the format of URLs
• outline how the Telnet protocol works
• outline how security features can be implemented with HTTP
• describe how FTP and TFTP transfer files work and the circumstances in which
they are most appropriate
• describe how remote files are accessed through NFS
• describe how DNS resolves hostnames to network addresses
• outline how SNMP manages remote network devices.
45
Appendix B: Model answers and hints
Appendix B
Model answers and hints
Chapter 2
a. True or False
i. TRUE.
ii. FALSE – they contain virtual circuit numbers instead.
iii. FALSE – this is transmission delay.
iv. TRUE.
b. See Section 2.3.2.
c. See final bullet point in Section 2.3.1.
d. Noise Power = 0.2 mW
Signal Power = 200 mW
SNR
= 200/0.2 = 1,000
= 10 log (1,000) dB
10
= 10 x 3 dB
= 30 dB
The noise will be amplified by the same amount as the signal (40 dB)
Let N = amplified noise
Noise Gain = N/0.2
= 10 log (N/0.2) dB
10
40 dB
40
= 10 log (N/0.2) dB
10
= 10 log (N/0.2)
10
log (N/0.2)
=4
N/0.2
= 104
10
N
= 0.2 x 10,000
= 2,000 W
e. Least number of links is a ‘daisy chain’ topology with five links.
The best way to improve resilience by adding one link is to form a ring.
Maximum resilience will be obtained from a full mesh.
Minimising delay to one node requires a star network centred at that node.
Chapter 3
a. True or False
i. FALSE – corresponds to top three layers of ISO Reference Model.
ii. TRUE.
iii. TRUE.
iv. FALSE – Service Data Units are passed.
b. See bullet points in Section 3.1.
c. Give three standards body names from Section 3.3 and find an example of
119
a standard from each (Note: IETF = Internet, ITU-T = Telecomms,
ISO = miscellaneous).
d. See third bullet point in Section 5.3.
e. See first paragraph of Section 5.4 and Figure 3.5.
Chapter 4
a. True or False
v. FALSE – this is IMAP.
vi. FALSE – it is inefficient and it runs over TCP.
vii. TRUE.
viii.TRUE.
b. See Section 4.5.
c. GAMES = 0110010011101.
d. See Section 4.6.2.
e. See Section 4.8.1.
Chapter 5
a. True or False
i. TRUE.
ii. FALSE – it is initialised to a number derived from the system clock.
iii. FALSE – it is the next sequence number it expects to receive.
iv. TRUE.
b. See first bullet point list of Section 5.2.
c. 10101100 10010101
00100100 11100100
10001000 01110001 = checksum
10101100 10011001 (errors in bits 3 and 4)
00100100 11100100
10001000 01110001
00000000 00001100
As total is not all 0s, errors have occurred.
d. See Section 5.4.
e. See Section 5.5.
Chapter 6
a. True or False
i. TRUE.
ii. FALSE – it’s the MTU Size.
iii. FALSE – it’s Class B addresses.
iv. TRUE.
b. See Sections 6.1.1 and 6.1.2.
c. 192.16.128.0 = 1100000 00010000 10000|000 00000000
192.16.135.255 = 1100000 00010000 10000|111 11111111
netmask
120
= 1111111 11111111 11111|000 00000000

Data communications and enterprise networking

Transcription

Similar documents

ASE2000 Version 2 - Applied Systems Engineering Inc.

Ilpuguasigm presentation

FS 1000 Owners Manual

ordering instructions mail order

Automating the hacking process

has royalty free clip art images to use for business cards or brochures.

We`ll help you get rid of junk mail! Please place unwanted catalogs

How to setup Windows Explorer for FTP in Windows XP

computers and systems

WS_FTP Pro Data Sheet - Secure Information and File Transfer